Reflections
50+ recent reflections
Monday, September 7, 2026
Five-day window spanning the Labor Day weekend. The through-line is authority placement - who owns a question, and who has to feel the cost of the answer. The largest decision of the week was refusing an SDLC and release-SLO contract in the Brian Dawson and Brady Dibble sync: engineering will ignore any such document, and the only thing engineering will not ignore is Product, because prioritization is the only real power Product has and every commitment engineering silently absorbs takes that power away without anyone deciding to. Peter proved the rule with the live case the same day - the CEO surfaced that the NVIDIA contract requires distributing the latest drivers within 30 days, and rather than assigning the work Peter filed it into the product queue while setting the one gate that is genuinely his: no further RLC-AI release without the latest drivers. On the commercial side he shook hands with Bjorn on a KT bridge - a year of CVE support on a non-LTS 9.8 against a migration commitment and a million dollars - and answered Nathan dislike of it with capacity rather than argument, because the money converts to one or two engineers who stay in Nathan org afterward. Two smaller calls ran the same way: he routed a candidate in two internal loops back to the candidate herself rather than let recruiting arbitrage her against req difficulty, and he closed an equivocal security-engineer read as a no rather than adding interviewers to it. He also extended a contractor under Wallace to keep the Atlas and Motherbrain applications alive after judging the code on its engineering merits, and set the target of collapsing nineteen engineering titles into four.
Wins
The prioritization argument landed as a demonstration rather than a lecture - Peter had already filed the NVIDIA ticket into the product queue before the conversation, so the principle came with an existing proof. The KT deal is a clean instance of the capacity rule doing real work: it converted a leader stated objection into a workable yes without relitigating the technical point. Wallace independently reached the same architectural read of the Atlas and Motherbrain code that Peter had, which is a good sign for the placement. Two decisions from earlier weeks came back with positive outcomes: the Google kernel-documentation hold cost nothing relationally and the relationship is now openly collaborative, and the prototypes-arrive-with-headcount rule held intact under pressure.
Challenges
The TPS report shows the cost side of the same story. Every top-three PIC priority is overdue - Arcee AI and Toyota Systems each ten days past target with confidence collapsing from 98 to 75 and 70 near the deadline - and the NVIDIA driver automation ticket now sits in Core Distro parking lot at 0 percent confidence, which is exactly what routing it to Product looks like when Product has not yet ranked it. The NARF service-accounts item lost 90 points of confidence in a week. Separately, Peter told Justin plainly that he sees step-change investment in Justin org and does not see it in the upstream Linux org, and that if the upstream does not evolve then nothing downstream evolves - a concern he has not yet turned into a decision.
Wednesday, September 2, 2026
A three-day window, Aug 31 through Sep 2 - eight decisions logged, five of them open. The through-line across the open five is jurisdiction rather than judgment: in each one Peter first settled what kind of question he was being asked, then answered only that. The Atlas code review returned an engineering verdict and explicitly refused the adjacent business question, keeping the bulk of the codebase, killing an interim laptop-resident PR and attaching a standing instruction that skills with direct data access stop immediately. The KT letter refused to let a sales document commit CIQ to a release that does not exist, and substituted the commitment CIQ can actually keep - 9.6 LTS now, with its real support window and a required migration - gated on engineering verification before signature. A partner indemnity escalation that arrived at engineering only because the blocker was technical was returned to its commercial owner in a single sentence, six minutes after the ask. At the engineering all-hands the write-static-code guardrail was clarified, a correct public objection was conceded out loud, and the policy held anyway on stated cost-benefit grounds rather than on being airtight. The one decision running the other way was an assignment - a field artifact request taken up as a real engineering deliverable, with an owning org and a scope line named in the same message. Three further decisions were logged sensitive and are not described here.
Wins
The KT letter was the sharpest work in the window. A seven-figure deal had a same-day signature deadline in another timezone and a draft letter that would have forked a release on paper. Rather than either signing it or simply refusing, Peter reframed the question first - why do they want early access, what are they actually trying to accomplish - then constructed the deliverable CIQ can honour and made it conditional on engineering verification before anyone signed. He then corrected his own support-window arithmetic in the group thread, unprompted, rather than let a wrong number reach a customer contract. The all-hands was the second win. A written missive two weeks earlier had produced confusion rather than compliance, so it was re-delivered live alongside the concrete failure that motivated it. When Jamin Collins raised a correct objection - an over-helpful agent will eventually route around any boundary that is not a real guard - Peter said out loud that he was right, twice, and kept the policy anyway on explicit cost-benefit grounds, inviting a better pattern rather than defending this one as complete. Third, the Atlas review closed a politically loaded question with only the part he was qualified to answer.
Challenges
Delivery risk is concentrating in the TPS numbers rather than in the decisions. Eight milestones are overdue and three sit blocked at zero percent confidence. The two large Fuzzball pre-production commitments are five days past target, with confidence dropping twenty-three and twenty-eight points as their deadlines approached. Errata scanner-vendor confirmations are blocked at zero. A FIPS kernel release is blocked with a seven-day slip, and FIPS meta packages for one LTS line lost seventy points of confidence in a single move. Separately, the largest partner opportunity remains blocked on something engineering cannot fix - Rocky is not a certified OS on the target hardware against a five-nines uptime contract, and the risk-transfer figure under discussion was roughly eight million dollars of guarantee. The Atlas review also surfaced a live production system running from a contractor laptop with unrestricted write scope into a system of record, bypassing the approval rules that same system enforces on humans. That is the failure class named at the all-hands, found in-house rather than hypothetically.
Monday, August 31, 2026
A weekend-plus-Monday-morning window: low volume, high concentration. Four decisions, weighted toward hiring and the Core42 engagement. On hiring, Peter decided the posture of the Jori Koolstra interview before the interview - he told Nathan Blackham his plan was to sell rather than evaluate, confirmed it within minutes of the Sunday call with Hire him now, and pushed for a roundtable inside 48 hours. On Core42, an overnight escalation asking CIQ to compress a five-week deployment produced two decisions inside one hour: hold the contracted five weeks while describing the padding honestly and the two-to-three-week landing as likely, and hand the customer urgency back as a condition by making speed contingent on topology detail only Core42 can supply; then route the deployment staffing into Ryan Smith org with Nathan capped at advisor and the unfilled Forward Deployed Engineer entered on the contract access list as a TBH. One further decision logged this session is marked sensitive and is not described here.
Wins
The Core42 timeline position was fixed before anyone had to speak to it externally - the joint Core42/NVIDIA call an hour later carried no date at all, only shorter than originally discussed, which is exactly the line Peter drew. The Aug 19 pre-emption of a three-week Core42 commitment is now confirmed to have worked: the SOW closed at five weeks and the compressed figure was never available to be quoted back under pressure. On hiring, Bjorn Hovland named the Rocky community-evangelism gap and said CIQ has no active community leader there, and Peter had a candidate for exactly that gap in the same 48 hours. A scope promise Peter had made to Nathan in a 1:1 held under real delivery pressure the first time Nathan tested it.
Challenges
One of the four decisions was forced by pressure arriving from outside rather than chosen on Peter own clock - a customer escalation that landed overnight and had to be answered before the joint customer call at 07:05. Separately, the engineering status report is carrying real risk in Performance Intensive Computing - Fuzzball pre-production for both Arcee AI and Toyota Systems went overdue on Aug 28 after confidence collapsed near the deadline, and three date slips were logged with no matching confidence update at all.
Friday, August 28, 2026
Four days in which almost every logged decision turned on where a question belongs rather than on what its answer is. The spine of the week was AI spend, and the shape of the call was not a number. Facing a token run rate heading toward roughly three million dollars a year with the top ten spenders accounting for about ninety-five percent of it, Peter refused a cap for the third time and refused a company-wide policy as well, directing instead that every overage case be diagnosed individually and answered with whatever that specific cause calls for, under a standing constraint that the answer may never be to suppress AI usage. He then gave four visibly different answers under that one rule in four separate conversations. When one of his own reports pushed for the AI committee to write a best practice, he conceded the conversation had to happen but took the venue away from the committee and put it in an engineering all-hands he will run himself, on the ground that the message was already arriving distorted through managers. The same move recurs across the rest of the week. In the exec product prioritization session he left the board with Product and contributed a criterion instead of a verdict, generalising a single reordering into a standing rule when he was asked to: something broken that we are shipping is an automatic fast path, regardless of the size of the epic around it. In a department heads thread he stopped a capable leader from volunteering a week of engineering work, not on the merits, but because the channel is not an intake path and the request had not passed through Product. Handed a cross-team tool dispute, he declined both to mandate adoption and to arbitrate the tool, diagnosed it as a communication failure in which neither side had said the true thing to the other, and put it on the managers agenda as a case study. Opening a code review he owes on a data system, he named out loud which half of the question was not his to answer, refused two offers of a use-case walkthrough, and asked instead for the scripts that still live only on one laptop to be pushed to the repo so he could read both ends of the system. And on the one capability gap he says he cannot close with the reqs he already has, he ran the partner path and the hiring path at once rather than queueing behind either.
Wins
The item flagged in Mondays reflection as overdue at zero percent and needing a real answer rather than another date change - the Pro Hardened kickstart-capable installer ISO - completed on 8/25. The same-version EL migration script shipped 8/27 at full confidence, and the backend MR release automation landed 8/24, three days late. Six older decisions got outcomes derived from this weeks activity rather than sitting unrated: four at four out of five, including the refusal to act on the spend number until it was decomposed, which is precisely what made this weeks per-case rule possible. The rule that an agent must call inspectable code rather than a system of record directly has now propagated outside engineering entirely on its own - it was restated to a team that does not report to Peter and came back confirmed as already-built-that-way, with no re-litigation.
Challenges
Both Fuzzball pre-production commitments are due today after their confidence dropped twenty-three and twenty-eight points on 8/25, and the service-accounts automation collapsed ninety points to eight percent. Two items are outright blocked, one of them at zero percent confidence pending scanner-vendor confirmations that are outside CIQ control. The harder problem is not the delivery risk but the instrumentation: the spend dashboard and the expense tool disagree by roughly eight times for at least one person, while the dashboard reconciles against the actual invoice - and governance by visibility only works if the numbers people are shown agree with each other. A related failure surfaced the same week: a strong individual contributor was approached separately by two different people about the same number and read a top-spender list as a naughty list, which is exactly the distortion the all-hands is meant to fix. Separately, a settled commercial fork had to be re-asserted when the team drifted back toward reopening scope that was already won.
Monday, August 24, 2026
A Monday whose decisions split cleanly into two kinds: one long-horizon instrument built with the CEO, and four short-horizon calls that each turned on refusing to pay a cost somebody else was about to incur on his behalf. The instrument is the engineering ladder. Peter took it to Greg first and deliberately withheld the name mappings, saying he wanted Greg bought into the structure and cared much less about the mapping for now - roughly 27 levels collapsed to four IC levels plus Technical Fellow kept as a title outside the ladder rather than a level, and leadership as either VP or Director with no Senior anythings. Greg objected to VP on optics. Peter did not defend the word, he defended the need - I wanted a senior leader and a junior leader title, what we call them I am far less interested in - and then traded: he let Greg define VP company-wide, so it spans organizations and reports to a C-suite with SVP meaning junior C-suite, and got engineering VPs in exchange, committing that Greg will never see him arguing for a VP who does not report to him. He named the cost he was buying out of explicitly, that once engineering has VPs he has to have conversations with everyone else about how you become one. The short-horizon calls run on the same instinct in different currencies. On the Core42 signing call he watched the team answer broad security questions honestly, gave them back the frame they had lost - ultimately there will be 2 SOWs - and then closed it: this is won, we should be careful not to re-open ANYTHING that does not need to be reopened, followed by stopping the screen share. Bjorn wrapped inside the minute and the final package went out that afternoon. In the Kubernetes sync he priced a runtime fork as a sales-cycle cost rather than an architecture question, arguing as a hyperscaler being handed a pile of validation hoops before it can sign, and landing it as a rule - it has to include existing rail widths, otherwise nobody is going to buy our trains - while leaving the door open that defaults can change once Fuzzball has traction. On AI spend he chose notification over caps, with the trust clause carrying the weight: if you do not know you are doing that, stop and figure it out, and if you do know and it is right, I do not want to get in your way. And on the new forward-deployed hire he took Bjorn word spearhead and rebuilt the role around it - a distributor of work who can tap a shoulder for 14 hours in somebody else org, whose job is to work themselves out of a job at each deployment - while flatly rejecting the premise underneath it, saying his confidence that the Core42 deployment is all-remote is zero, absolutely zero. Two smaller calls close the day: he backed Stratum as the product name while saying out loud that he rejects the reasoning behind it, and he kept a reference requirement that was about to be waived on the grounds that the reference itself does not matter but not being able to produce one does.
Wins
Three deliverables completed since the Aug 21 report, all landing today: the NARF backend MR-to-release automation at 70 percent confidence, RLC Pro AI 9.8 taking CLK 6.18 as its default kernel at 82 percent, and RLC 9.8 standalone bare-metal and VM install artifacts at 90 percent - the last of which clears one of the three items that were sitting seven days overdue at the previous reflection. Two prior decisions converted and were rated four. The Aug 13 decision to take the re-leveling from principle to a concrete ladder reached CEO ratification of the structure today at a stated 85 percent alignment. The Jul 24 approval of plus-one US headcount for Ryan to build forward-deployed engineering in-house turned into a specified role with a JD in first draft and a named candidate identified. Core42 SOW1 reached the sentence the whole effort was aimed at - can we sign this - and the final package went out the same afternoon for conversion to Core42 template and CEO signature, with hardware arriving around September 18 and deployment starting days after. The NARF system and service accounts item that collapsed ninety confidence points on Aug 21 has recovered from 8 percent to 30 percent against its new Sep 4 date, and the same-version EL migration script sits at 100 percent in acceptance testing against Aug 28. Three derived outcomes were confirmed this session.
Challenges
RLC Pro Hardened and Pro Hardened LTS shipping a real kickstart-capable installer ISO is now overdue by three days at zero percent confidence, with the ITPIE customer ask behind it. This is the fourth consecutive reflection to name it and it has still not moved a point; today it was pushed to Sep 4 and then pulled back to Aug 21 on the same day, which is date churn rather than progress. Fuzzball support for the AMD evaluation is now overdue by ten days, up from seven. Errata scanner-vendor ingestion confirmations remains blocked at zero percent with no target date at all, and CLK 6.18 FIPS is still blocked at 47 percent against Aug 28. RLC Pro/Plus 10 slipped twenty-one days, Aug 14 to Sep 4, at 89 percent. Two exposures come from the decisions themselves. Peter told Bjorn and Ryan that the cloud number is 100 percent Fuzzball and to get used to it as the new normal, on the strength of an analysis he has not run yet - if the split turns out to be development spend rather than customer-passable, that guidance is wrong and Bjorn will have modelled against it. And the sales-engineer demo structure ratified on Aug 11 is holding in practice, but the enabling work it depends on sits in the parking lot at 50 percent against Sep 15, which is why that outcome rated a three rather than a four.
Friday, August 21, 2026
Two days that end on a Friday in which five of the six decisions on the record are aimed at one thing: the throttle on the Linux lead. Peter worked it from four independent directions in a single day rather than through one conversation. He absorbed the Core42 Dubai build-out outright - CIQ has no sales engineering function, so if something gets built at this company engineering builds it, the resources report to him, and it is 100 percent a Peter problem rather than a Nathan one, with the stated goal that Nathan never spends another brain cell worrying about it. He removed the gate Nathan had communicated to HR, refusing the fill-every-role-first version of the substitution and shaking hands instead on a one-for-one trade: yes to the Nikola Kolev hire, against Nathan getting aggressive on the departure it is meant to enable. He re-cut a solo RESF contribution as an institutional ask - the RESF asked CIQ, not Nathan, so the secure-boot request goes to all of Linux Engineering in case one person raises a hand. And he changed what he reviews: the metrics conversation moves off the tooling and onto the whys, because pounding the drum on pace gets him nowhere and what gets measured is the lever that does work. Underneath all four sits the diagnosis he finally named - Nathan is not less aggressive because of ability but because he has never had a startup collapse underneath him - which is also why he re-anchored his own no-more-Amazon-blood rule on the deficit it was actually pointing at rather than the company it named. The sixth decision runs on a different axis but the same instinct about scale: abort-the-loop authority drops from the hiring manager down to every individual interviewer, because at this size he would rather trust individual judgment than buy consensus.
Wins
Two deliverables closed since the Aug 19 report, both on Aug 20: the Partner Portal MVP engineering delivery at 81 percent confidence and the Portal migration from HubSpot Payments to Stripe Checkout at 95 percent. The absorb move validated itself from outside - the Core42 HLD final review closed on the line that implementation is coming soon and that is a Peter problem, said by somebody else, matching the conclusion Peter had already reached privately in the Nathan DM the day before. The metrics decision converted inside the same conversation it was made in: Nathan volunteered to start putting the why onto the dashboard himself, and Peter added the second reason, that it stops people gaming a metric through misreading why it is tracked. Four derived outcomes rate a four. The five-week delta requirement from Aug 13 is visibly operating - Steve has an engineering ops scorecard in front of his team that is already changing behaviour and cleaning up Jira, with Norm giving useful feedback on it. The Core42 solution-review redirect toward whoever watches the recording also rates a four, since the engagement advanced to implementation rather than being disqualified. Releasing the 1:1 tracking mechanism to Nathan while holding the measurable-outcome requirement held on both halves. And the Patrick Culp offer, made US-only with no forward-looking statements, rates a four: he is on staff and is now one of the engineers Peter counts when sizing Core42 build-out options. Seven derived outcomes were confirmed in total this session.
Challenges
RLC Pro Hardened and Pro Hardened LTS shipping a real kickstart-capable installer ISO has now arrived at its target date at zero percent confidence and still in progress, with the ITPIE customer ask behind it. This is the third consecutive reflection to name it and it has not moved a point in that time. A whole cluster of other Aug 21 targets was pushed in the days immediately before the deadline rather than earlier: Remediation Automation system and service accounts for Peridot-to-Release had its confidence collapse ninety points, 98 to 8, on Aug 21 itself and then moved to Sep 4; automated package-set validation slipped fourteen days at 52 percent; Fuzzball v4.3 turn-key AI models slipped twenty-one days at 20 percent; the Fuzzball Homebrew CLI item slipped fourteen days at 20 percent. A same-day ninety-point drop is precisely the failure mode the confidence methodology set two days earlier was aimed at. Errata scanner-vendor ingestion confirmations sits blocked at zero percent with no target date at all, CLK 6.18 FIPS is still blocked at 47 percent against Aug 28, and three items are now overdue by seven days, up from five: RLC Pro/Plus 10 and RLC 9.8 standalone bare-metal, both in acceptance testing at 89 and 90 percent, and Fuzzball AMD evaluation support. The shape of the decisions is itself the challenge. Four separate instruments in one day aimed at one leader, resting on a diagnosis that neither scope absorption nor a hiring filter fixes on any short timescale. Peter said out loud that pounding the drum gets him nowhere, which is honest, and the replacement instrument is a conversation scheduled for the next 1:1.
Wednesday, August 19, 2026
A twenty-five hour window running from the afternoon of the August board meeting through the middle of SKO week, and almost every decision in it is about what a number or an artifact is allowed to claim. Three separate items are the same move in different domains: bake the confidence figure on the scope in front of you today and let it move only when scope moves, refuse to act on the AI spend total until one-time hardware is separated from recurring API, and turn the five-week leader presentations from a recurring dashboard read into a standing requirement to show what you changed to move the number. In each case Peter defined what the instrument is for and then refused to let it absorb risk it should not carry - and in two of the three he named the cost of that out loud in the same breath, predicting that a ninety percent will slip and that he will have to explain scope changed to Bjorn. The other cluster is Core42, where the whole effort was aimed at not letting a widening conversation eat a near-term close: fork the engagement into two statements of work and let nothing delay the first, hand Bjorn the real setup sequencing privately so a three-week delivery number dies before it is said in front of the customer, and show the responsibility matrix filled in as a flexible first cut rather than arriving with it blanked. He conceded every cosmetic change asked of that document - branding out, logos out, draft stamped across it, the aggressive phrasing softened - and refused the only one that would have removed content. Underneath all of it sits the general form, delivered to his leaders the day before: value delivering things over doing work, ship three items all the way out the door rather than move nineteen forward.
Wins
Four deliverables closed on 2026-08-18, including RLC 9.8 Pro H - one of the four items flagged overdue the day before - along with Fuzzball v4.2 MCP access for AI agents at 98 percent confidence, the Remediation Automation front-end MR pipeline, and Warewulf Pro v1.3. The first-cut call on the Core42 responsibility matrix converted inside eighteen hours: it went in front of the customer as a draft the next morning and became one of two named Thursday deliverables. Six derived outcomes landed this session, three of them at four or five. The stalled GB200 order rates a five - the chassis is bought, and the over-utilisation complaint Peter pre-absorbed when he forced it through arrived exactly as predicted through the finance AI-cost line, where he dismissed it as one-time rather than relitigating the purchase. Letting Brady raise the kernel-team resource block himself rather than carrying it rates a four: Brady raised his own cost-visibility ask unprompted in the weekly sync and got Wallace to commit dashboard links. And the value-driver doctrine is visibly operating - the entire sync ran off target date plus confidence per item with leaders arguing their own numbers, and Wallace reported migrating more of his team work onto the NGD board.
Challenges
The delivery hole named in the last reflection has not moved and is now two days from its target. RLC Pro Hardened and Pro Hardened LTS shipping a real kickstart-capable installer ISO still sits at zero percent confidence against 2026-08-21, with a named customer deal behind it. Three items remain overdue by five days, and CLK 6.18 FIPS is still blocked at 47 percent after slipping to Aug 28. Peter is addressing this class of problem by rule rather than by intervention - the confidence methodology and the do-fewer-things-all-the-way message are both general instruments, and neither of them moves a zero percent item due Friday. The Core42 delivery boundary is also leaking: the 2026-07-30 decision that Core42 delivery would not land on Nathan org rates a three, because three weeks on no purpose-built team exists, the linked todo is unactioned, and while the deliverables are being carried by CIQ Federal and Dawson, Nathan is recording debriefs, is named as required on the control-matrix review, and Bjorn offered training on Nathan team as needed.
Tuesday, August 18, 2026
Five days in which almost everything Peter decided was about routing rather than content - who owns a question, who a message should come from, and which number an argument is allowed to turn on. The largest item was the first-ever reversal in Rocky usage, which arrived the morning of the August board meeting with a ready-made explanation attached (two departures to Alma). Peter refused the explanation on magnitude grounds and refused to supply a competing one even when Greg asked for his gut, converting the agenda instead into a question about whether the count-me figure is the number CIQ should be tracking at all - a dichotomy where either answer commits someone to something. In the same conversation he asked Greg to take him off the RESF directive Greg was about to send, having worked out by simulating one recipient perspective that including him inverts the message from CEO-tells-RESF-to-ask into CEO-corrects-his-CTO; he took a copy instead of a seat, and conceded that his own standing posture of waiting for RESF to ask has not worked. He also settled a data-classification question by refusing to answer it per-system: the policy already existed, the only unclear fact was whether HubSpot counts as customer data, and new AI systems inherit the Claude envelope rather than negotiating their own. And he held a hardware queue closed - a DGX Spark deferred a few weeks with a named path back in, and the reasoning carried to Bjorn personally rather than through an assistant. Two additional decisions from this window are logged sensitive.
Wins
Two derived outcomes rated at the top of the scale or near it: the Google lane routed to Kelly Hall is now being run independently, with Kelly declining committed-use discounts on her own margin analysis, and the honest slower live-kernel-patching timeline cost nothing - Tissa is still engaging and the next engineering call is being scheduled. The AI-data hygiene rule from 2026-08-11 also held its first real test, extending to two new systems without forking and without becoming a per-system approval. On the engineering side, RAT completed the front-end remediation-automation deliverable on 2026-08-18, Warewulf Pro v1.3 with the kdump overlay shipped, and Nutanix expanded 2x to 300k over three years while dropping the weekly custom build that had been a standing tax on the team.
Challenges
The RESF situation is the hard one, and Peter said the quiet part himself - parachuting did not help. Holding the line that RESF must ask has produced neither requests nor progress, and the download reversal removed the room he had to keep waiting. The fix he chose is indirect on purpose: pressure applied by the CEO from a direction that preserves his standing, plus a standing status cadence with Lee because Lee has the broad view and does not report to him. Whether that produces visibility is the open question. Separately, the NARF ownership decision came back mixed - one deliverable shipped, one lost 12 points of confidence and slipped past deadline, one is parked at 46 percent - which is the first real evidence that naming an owner with a cannot-wait instruction is not sufficient on its own. And the engineering delivery picture has a genuine hole: the RLC Pro Hardened kickstart-capable installer ISO sits at 0 percent confidence against a 2026-08-21 target, with a named customer deal behind it.
Thursday, August 13, 2026
Two days in which the biggest structural change Peter has attempted all year stopped being a spreadsheet and became a set of conversations. On Thursday he walked Nathan, Steve and Ryan each through the leveling map individually - roughly 13 engineering levels collapsed to four, with Nathan arguing him toward a fifth, and his own leadership layer smashed down to Director and VP. Ryan was told his bucket in the meeting. The constraints he set are the tell: nobody gets a pay cut, options cannot legally be clawed back, Mariah has to buy the framework before any org sees its mapping, and the whole thing lands in weeks not months. He was explicit that the ladder is not the point - it is the instrument that finally makes a compensation question answerable, because you cannot say someone is underpaid relative to peers when a Director with no reports and a Director running an org occupy the same nominal rung. The same instinct ran through the rest of the window. On the five-week org reviews he was indifferent to which metrics a leader picks and immovable that the second presentation must show the delta and what they did about it. On the on-call question, brought to him after the Portal outage with the remedy already chosen, he declined to start from the remedy and required an SLA first, keeping accept-the-downtime on the table as a real option. In each case he refused the specific proposal and restated what the artifact is for, then let the answer fall out of that. Core42 got the sharpest version of it. After a morning call where he was genuinely happy with the team, he concluded the room is not the decider - the security organization is married to a Red Hat solution and is mining the calls for disqualifying gaps - and redirected the entire effort toward whoever reviews the recording afterward. Less detail, not more. Every answer technical, never personal. He priced it at 10 to 15 percent and took the shot anyway, because Core42 engineering has already put CIQ over Red Hat, Canonical and WWT in writing, and the deal is Stargate-scale. Hiring showed the same judgment applied twice with opposite results in a single week. Leandro was compressed into one week on the strength of a rare like-mind endorsement from Nathan, with Peter running the final interview as a sell call and confirming two hours later that he was a yes. In the Wednesday round table, facing a candidate nobody was excited about, he deferred entirely to the technical panel, declined to block, and attached a stop condition - if he does not accept, do not kill ourselves - before closing with the line that mattered most: I know this team has felt pressured, and I do not want this pressure to bring in the wrong person.
Wins
The leveling work crossed from theory into execution - three directs walked through their own mappings in one day, with the funding constraints and the sequencing both stated up front rather than discovered later. Hailey Mothershead drew an unreserved hire-today verdict and offers were scoped in the room. Leandro was moved from Monday interview to Friday decision call inside a single week. The Core42 engineering organization has CIQ over Red Hat, Canonical and WWT in writing, which is a real result regardless of where the security review lands.
Challenges
Core42 is a deal Peter believes CIQ has won on merit and will probably lose on politics, and he had to hold two things at once - praising the team performance while telling them privately that the outcome is roughly 10 to 15 percent. The kernel team is carrying open reqs and visible pressure at the same time, which is exactly the condition under which a hiring bar quietly drops. And the TPS report shows the problem the new review format is designed to catch: several items slipped between 63 and 137 days on a single day in early August without a confidence update.
Tuesday, August 11, 2026
Twelve days, with COVID taking most of one of them out, and the period is bracketed by two things that could not be less alike: the first major Fuzzball deal closing, and a rule about function calls finally landing. The rule came first. Friday at the AI Committee he re-delivered the thing that had failed silently a week earlier, and this time he did not assert it - he demonstrated it, walking the room through his own tool to show that AI helped create the capability but is not part of the runtime function of accessing anything. Claude cannot get to Slack, except through that function. It does not have the keys. The ask was exactly one sentence long, and he declined the two obvious extensions on purpose: no shared library, because there is no value in building a library of these things and forcing an MCP server would block people from experimenting with technology being stood up today, and nothing retroactive, because a year from now we are going to have 100 times as many tools running around the company. He gave three reasons rather than one - inspectability, 100 percent repeatability, and teaching the company the right pattern, since Claude is going to make everybody at the company a developer and that is incredibly terrifying if we do not teach people to be better developers. Then the part that made it land: Michelle told him people cite his laissez-faire stance to bypass her review, he said I know I have set that up, and offered to go tell the company himself that this is not a roadblock. Five days later Ryan had filed a Jira epic turning it into a Customer Engineering department standard with a generated exhibit and two reference implementations, and had audited his own Mission Control against it. That is the rule being repeated back in the strongest form available. The Arcee close he handled by redistributing the credit in writing and then making sure everyone saw the redistribution. To the Fuzzball team: Bjorn pounds his chest about getting the deal across the line (and he earned that - he was hounding them daily). But you are the reason there was a deal to close. They did not buy a pitch. They bought Fuzzball. He named the long nights, pre-forgave the deployment quality - Arcee is the first of these, not the last, and it is the worst Fuzzball deployment that is ever going to happen - and credited the pace itself: finding a new gear is one thing, staying in it is the hard part, and you are doing it. Three minutes later he reposted the whole thing into department-heads with just so you all have visibility into what I am telling the Fuzzball engineering org, which is what converted a thank-you into an organizational statement about what causes a deal. He had told Chris the same thing that morning and asked what channel would reach the whole team. Google ran through the whole window and he played it the same way twice: protect the relationship, price the work honestly. When Tissa reported that CIQ wanted to delete the RL 8.6 source tree, he replied the next morning without consulting anyone - I do not know why my team informed you they wanted to delete that tree. I apologize. There is little cost associated with us keeping it live. I hope it was presented as a discussion, not as CIQ is doing this - and took the internal question offline, where he separately flagged the motive he suspected and rejected: I do not want to do anything that they are nervous about right now, especially if the reason is that its existence makes our metrics look bad. I want to make sure we have the actual contracts completely signed in blood. Ten days later that goodwill produced a paid live-kernel-patching quote request, and he opened it to Nathan and Justin with the escape hatch removed - assume we get headcount to cover this work - interrogated the shape rather than the number, landed on 2 HC for 3.5 months and 1 ongoing, did the arithmetic against Googles October date out loud, concluded I think that gets it to them too late and they will say no, and sent it anyway. Lets go fishing and see what we can catch. Third time this instrument has run in three weeks after Rakuten and Citadel. The LG recovery was built the same way as a purpose statement. After sales delivered a mea culpa that they had been pitching solutions that do not work and did not understand the problem, he stripped the next meeting down to one function: 100 percent about understanding their problem. Not telling them no, we will not do this. Not telling them no, we cannot. The response gets written afterwards with walk away named in advance as a live option, and he told Justin explicitly that the signup is not, hey Justin, look what you get to build - while pre-declaring the line he will hold once a response exists: if we provide it, we are giving some surety that it is supportable and that it will work, so I want to make sure we are staffed for that. He did the same thing to the product-prioritization argument, which had been running in a Jira thread for five days: Jira (and slack) are not conducive to good process debates. Jira is a great system of record. That is all. He closed the thread, decided nothing on the merits, and named the actual harm - the ripples this thing is creating across the company are substantial and unnecessary, all on the same team, all trying to get to the same place. The hiring calls all turned on refusing to promise or refusing to weigh. On the Japan relocation he cut Nathans already-cautious framing by another twenty percent - I would not make any future looking statements, even something, I would not make any future looking statements, period - authorized present fact only, anchored against the candidates own Amazon background so the no reads as a difference in company stage rather than unwillingness, and accepted the cost out loud: and if we lose him, we lose him. On the kernel candidate he disqualified his own reaction before hearing anyone elses - I was not in love at all, but I think my issues are just personality mismatch, which I cannot allow myself to weigh heavily - then reclassified the real objection from the person to the role, declined a 1:1, and placed it in the group conversation with Brie because that is where the req owner would hear it. Usually this talk with Bri is a rubber stamp. I am looking for this one to be more of a discussion. And on internal moves he answered the rule rather than the case, anchoring on a real CIQ comparator - I would like to be very careful setting a precident that moving within the company gets new comp - which he later generalized further: it is always about understanding precedent, whether it is a comp change or anything that structurally affects the company. The two calls where he sided against his own engineers were the most interesting. On prototypes crossing over from Product he backed Bjorn against both Nathan and Justin, and did it by repricing rather than by authority: it is a prototype, nothing more, the amount of code there is not more valuable than two days, so all we are burdened with is what we have learned in the market about what people want from it, which is an awesome thing to be burdened with - instead of Brady woke up one day and had a brain fart and decided that this is what a product should do. He named the changed variable rather than pretending the old objection was wrong, said out loud that it is a very different model than anything we would have done at Amazon, and then removed the capacity complaint three times with period attached: Ascender with headcount, Everfox desktop with headcount, Kubernetes with headcount. He pushed the logic one step further with Bjorn - release Satellite today with a go-to-market and no software, and see who rings our front doorbell. And on NARF he took the recommendation while rejecting its logic in the same sentence, saying he had trouble with the linkage of crowning someone partly because they had not moved the work forward, then converting it into four clauses for Justin: he has to own this thing, run it like he owns it, he has the authority to do what he wants with it, and he cannot wait on anyone. Three days later the org-wide note made it a fact. Two smaller ones carried real weight. He forced a stalled GB200 order to move the same day with the constraint restated rather than relitigated - as stated earlier, time is the important factor here, not dollars - because his read of the NVIDIA relationship is that utilisation buys more capacity, and he pre-absorbed the only downside by telling the committee to use it all you want and to send him the complaint when Scott notices. And he ratified a sales-engineers-own-first-pass-demos arrangement that a vacation collision had accidentally produced, telling the person who reported it as an error that they had kickstarted the right structure anyway - which is the second observation of routing non-differentiating work out of engineering rather than staffing it there. Five decisions this session were logged sensitive=1 and are deliberately absent from this summary.
Wins
The AI-access rule landed on the second attempt and was repeated back within five days as a filed department standard with a generated compliance exhibit - not agreement, but reproduction, which is the only real evidence a rule took. The Arcee credit note redistributed causation publicly and cheaply, naming Bjorns contribution honestly enough that the correction reads as accurate rather than consoling. The RL 8.6 apology cost a few dollars and preserved a relationship that returned ten days later asking to pay for more work. Twenty-three older decisions got outcomes derived from real evidence rather than guesses, including three fives: the inspectable-code rule, the standing benchmark-publishing permission that put CIQ within one percent of best single-node on Spark Arena, and the Google ELTS pay-or-I-stop call, whose agreement closed on Jul 30 and whose relationship then expanded.
Challenges
Two decisions came back rated 2 for the same reason: he ruled from first principles and the ruling did not take. The two-boards re-rule needed two further interventions inside a week, and the confidence-must-rise-as-the-date-nears rule is being violated in the live data - CLK 6.18 FIPS dropped 23 points near deadline, 70 to 47, and there are four separate date-slipped-without-confidence-update flags attributed to Nathan and Justin in todays report. That is the same instrument he defended so strongly two weeks ago now failing on the mechanics rather than on the principle. The Linux org state-of-the-state session also never happened - the invite lists for two teams silently did not work, only three people showed, and it was rescheduled into the week COVID took out. And the metric-design ask delivered a doc but not the handoff: I had wanted to feed Nathan, here are the metrics I want you to start tracking against. We did not.
Thursday, July 30, 2026
A 1:1 day, and almost every decision in it resolved the same way - by restating what an artifact or a process is FOR rather than arguing the specific case on its merits. The sharpest instance was the confidence column. Bjorn carried Bradys push for a uniform cross-team definition, and Peter refused the premise outright: confidence is a human thing, that is why we call the column that, it is not computed immutable expectation of delivery for a reason. He would not let the numbers be normalised because normalising them would strip out the only thing that makes them worth having - a senior engineers four-second judgment, kept current daily near a target date. He named the second-order cause too, which is the part a formula could never fix: Brady is WHY someone like Justin will not put high numbers down, because if you are working with someone you feel is constantly looking for things to say you were doing wrong, you rapidly start communicating lower numbers. The translation burden was placed on the consumer - if Brady wants ranges or high/medium/low in his own process, that is him translating, not Engineering fitting an arbitrary mapping - and he gave Bjorn an empirical test rather than an argument: ask Duane if he can work with the numbers Justin is providing, the answer will be yes. He closed the same instrument from the other end with Nathan, declaring the value-driver bottom row the contract and NGD merely a backing store - the reason I am using that language instead of NGD is because I do not care about NGD - then approving the mechanical change that follows, stripping target dates and confidence out of product tickets so they point at NGD instead, with only products desired date left behind. The same move ran through the people decisions, in the release direction rather than the defensive one. The 1:1 tracking spreadsheet became Nathans choice - I suggest yes, but I do not mandate yes - with two properties held back as non-negotiable: whatever replaces it must be measurable, and it must produce a record that does not go to HR so it stays open and honest and unpolluted. Steve got the identical latitude the same day: what I care is that there is something documented, whether it is the spreadsheets or not is not the important point. Both are scaffolds with an exit condition rather than permanent compliance, and in both cases the release came with a collection attached - if Nathan builds something better, that is what he presents at the Tuesday sessions. The engineering re-leveling got its delivery mode settled on the same principle: a spreadsheet benchmarking each org against all of engineering on both level and comp, walked through individually next week, and pre-emptively stripped of authority three separate times - it is the beginning of a conversation, not for this is what I am doing to your team. He delivered the Wallace-org diagnosis to Steve directly rather than about him, and paired the hard finding with its counter-evidence: titles far above the rest of the organization, but comp not high and possibly low, so you are going to have a fine story to tell. The remedy space was left genuinely open - up-level everyone or strip VP titles, both named as live. The hiring call was the days clearest use of delegated authority. Peter declared Nathans user-space reqs the highest priority in the company outright, then handed Nathan the diagnostic rather than intervening himself: be really comfortable asking Brie whether your recs are her highest priority, and if the answer is no, that is a conversation with you, me, Brie and whoever elses rec she thinks is higher. Two weeks had passed with the user-space side untouched, which made the priority question empirical instead of theoretical. On Core 42 he declined an offer of ownership twice - Nathan volunteered to own the deployment team and Peter judged the fit wrong rather than the willingness insufficient, routing it to Bjorns forward-deployed structure, naming sales as the eventual home, and flagging that the team likely wants to sit in the Middle East and fly to the UAE and Saudi from Qatar. The point of settling placement pre-close is so the account team stops promising capacity that does not exist. Two smaller calls carried the same shape. A hundred-dollar KVM purchase turned into a policy fight, and Peter split it cleanly: unblock Nathan immediately and absorb the risk himself - tell them No and that the CTO said it was fine, I do not want stuff like this to continue - protect the part-time contractor executing the policy from an escalation she could not resolve anyway, and take the actual argument to Finance as a threshold regime rather than an exception. If it is below your threshold and you approved it, I do not want it routing to me, I do not want to know about it, I never want to see it. And on C3 he answered within two minutes at night: no engineering owner, content wrong and outdated, manage it or kill it, and I do not care which - the same delete-rather-than-reconcile test he applied to board columns yesterday, with the indifference stated out loud so nobody has to come back to him for a preference. The counterweight to all the process work was Mondays session with the Linux org, which he fenced off hard. Gregs skip-levels showed Nathans and Justins teams measurably less optimistic than the rest of the company, and before acting on it he audited the instrument - watched recordings to check Greg was not leading the witness and asking the question the same way each time. The response is information and credit, nothing else: the Middle East picture from his second trip, the Axiom Space meetings, and the Google fight attributed explicitly to their groundwork - I got to do that because of the work your team has done, and I want them to hear that, I want to thank them for it. Universally positive. Period. Full stop. Fixing things goes in another meeting, because muddying the message would convert a recognition meeting into a performance conversation and the information would stop landing. He also committed to showing up personally at Fridays AI committee to re-deliver a rule he had already set once and watched not land - AI writes the code, but what data a dashboard can pull is hard-coded in the function, so an agent cannot wake up on a Tuesday and pull columns nobody expected. His own account of why it needs re-delivering was flat: I am happy making that clear to the AI subcommittee, and I thought I had. Two additional decisions were logged sensitive=1 and are deliberately absent from this summary.
Wins
The Google agreement closed - Tissa Senevirathne confirmed closure of the agreement by email with Kelly Hall in the loop, and Peter had told Nathan earlier the same day that CIQ has it in writing from Google on both the GDC LTS side and the GDC expansion that they will sign, with the GDC money backdated. That validates the 7/21 decision to pull the business side into the FIPS 6.18 call, now rated a 4, with CLK 6.18 FIPS for Google sitting as Core Distros number one at 70 percent and in progress. The 7/09 decision to pass on a candidate without a black mark and re-engage in three to six months converted exactly as designed - Nathan pulled the trigger inside the window and Peter cleared it with nothing further needed from him, rated a 4. The TPS report also shows real completion volume in the last two weeks: five Fuzzball items including the DGX Spark work and the PostgreSQL 16 migration, HR onboarding and offboarding automation, and two items at 100 percent confidence ready for release - the same-version EL migration script and out-of-the-box hardening for 9.7. And Nathan bought thirty coins for his team reading failure is not an option, it is mandatory, which is the culture of fast iteration propagating without anyone being asked to propagate it.
Challenges
The 7/21 decision to pause the AI Data Management policy and route it to Greg has to be rated a 2. It is still unapproved, it stalled while Greg was at a conference all week, and in the gap the Atlas project produced exactly the failure it would have prevented - an agent-driven dashboard pulling data nobody had vetted. Peters own read of why is the uncomfortable part: he believed he had already communicated the rule. That is now the second consecutive week in which a structural change failed on first telling rather than on substance, and the fix is again him re-delivering it personally rather than anything structural. Confidence hygiene has not moved either. NARF Peridot to Customer is now 41 days overdue at 80 percent confidence and Kernel CI Automation 2.5 is 34 days overdue at 60 percent - both one day worse than yesterday, which is to say untouched. GCE ARM Mellanox CX-7 RDMA is blocked against a Jul 31 target at 50 percent, and the Nvidia DGX Spark demo still carries an 80 percent against a date that slipped fourteen days with no confidence update. This is precisely the pattern Peter called out at Tuesdays sync, and it survived the callout. Meanwhile the Linux org reads less optimistic than its peers for the second consecutive week, and a peer function spent the week trying to standardise the confidence column - so the same instrument is being under-maintained by its producers and over-specified by its consumers at the same time.
Wednesday, July 29, 2026
A 45-hour window dominated by one escalation, one confrontation and one structural fight, all three handled with the same move - install the boundary or the criterion before arguing the substance. The LG U+ escalation matured from last weeks definitive no into a documented position. Against Howards feasibility document Peter opened with three flat lines - no CIQ packages presented as RH, no CIQ repos pointing at RH keys, no RH files hosted - and separated technically possible from business possible before touching any engineering, on the grounds that engineers had been loose with language and there are a bunch of ways to make this work technically while none that survive signing as CIQ without masquerading as RH. He built the escape hatch so it retires only the legality objection and not the cost objection: a lawyer plus evidence CIQ would not be a lone actor, and even then it is STILL another release we need to support. Separately he killed the reuse business case by attacking the amortization premise rather than the estimate - different customers want different packages, so the testing matrix multiplies, and the alternative is shipping untested artifacts, which he classified as a business failure rather than an engineering shortcut. And he set an information-hygiene rule for the escalation itself: if the team does not know how Oracle does it, we do not know is the right answer, be truthful but do not lead the witness, and the call belongs to Engineering and Product rather than Support - a standard he applied to Howard as well as to Ryans team. The Google confrontation was the weeks sharpest act. Facing an attempt to relitigate a February technical decision as grounds for withholding payment, Peter refused the retroactive-attribution frame on principle - technical decisions get made by the team and recorded as the consolidated teams decisions - then bought the pretext off, offering to absorb the disputed 150k against an existing 250k concession so that the only question left was binary: can you pay us, and if you do not, I am done doing work. His reasoning was that 150k is a rounding error to the counterparty, so if they let CIQ walk over it there is something we do not know and CIQ was always the backup plan. He priced the downside honestly - a logo loss and an unhappy president - and accepted it. Googles side stopped the meeting and committed to closing out within a day. Underneath the money he was defending a working condition: conceding the frame would have meant every future engineering discussion had to be paired with a does-this-impact-our-contract check, stripping engineers of the freedom to pick the right solution. The structural fight was the boards, and it is the weeks clearest failure as well as its clearest doctrine. Seventeen minutes of the Engineering Weekly went to re-deriving both boards from purpose - product priorities is prioritization only, the engineering deliverables board is confidence numbers and target dates for go-to-market-linked work only - including correcting a peer executives directive in the open and instructing that any column serving a third purpose be deleted rather than reconciled. He subordinated the tooling explicitly, and the following day extended the same logic downstream: if the report does not show a teams work, the report changes and pulls from two sources, but the board stays the same. He also tightened the confidence rule - near a target date a 60 percent is itself the failure, because the company cannot act on it, so the date should have moved earlier. Alongside that he owned the adoption failure without conceding the substance: the messaging around it is wrong, he has not had this much trouble making a structural change in 20 years, and what people are actually asking is how his process solves their problem, to which the answer is that it does not. The constructive counterweight was demonstration rather than specification. He restructured the staff meeting into a fast breadth-first pass plus a single-org deep dive, put his strongest manager first and said the ordering was deliberate, praised the thinking rather than the artifact, left the form open - dashboard or just a system that reports metrics - and then named the ask: this is what I am going to be looking for from every org. He propagated the recording sideways as a training artifact and privately calibrated the bar by warning that the rest of engineering will not measure up. He asked for the half-migrated report to be shown even if not perfect, which is what made the board divergence legible to the whole room at once. On Axiom Space he took the Fuzzball pitch himself when a relationship call turned out to sit inside CIQs own domain, set the path at NDA then contract, and named the window - they are just starting, so they are in a position to choose OS cleanly right now. He committed to a Monday storytelling session with the Linux org after Gregs skip-levels showed them meh on the future while two peer orgs are bullish, reading the gap as an information failure rather than a conditions one and bounding the response so it stays a session rather than a program. And he granted latitude rather than review on external open-source positioning, closing a future option in the moment with we will want it. Three further decisions were logged sensitive=1 and are deliberately absent from this summary.
Wins
The Google payment escalation produced movement: the counterpartys side stopped the meeting and committed to closing out within 24 hours, and confirmation followed that the ELTS PO is being issued. The 7/21 decision to hold the 6.18 image as leverage is now rated a 4 on that evidence. Axiom Space moved from an introductory call to an NDA in motion with a stated intent to close in weeks, off a pitch Peter delivered himself. Wolfords deep dive landed exactly as intended - Peter called it perfect and exactly what I wanted to kick that all off, and it is now the reference implementation for the rest of engineering. The Rocky team turned an LPE disclosure into a shipped patch in two hours forty minutes, beating Alma, and Peter recognised it in channel. Customer Engineering closed out Deep Support, PS Operations Automation and the training program, earning the in-house forward-deployed bet a 4.
Challenges
The operating model codified on 7/21 has to be rated a 2 - it did not take, and Peter said so himself: almost daily somebody is using these boards for something different than what he said, and he has not had this much trouble landing a structural change in 20 years. He accepted the messaging as the failure point rather than the audience, which is honest but leaves the fix unresolved. Confidence hygiene is failing across the board in exactly the way he called out - the fresh TPS report shows a demo slipped fourteen days with no confidence update, two Core Distro items more than a month overdue, and several near-term targets sitting in the fifties and sixties. The LG escalation consumed a large share of executive attention across three days for a question that was already answered on 7/27, which is the cost of untraining five years of saying yes. And the Linux org reads as meh on the future while two peer orgs are bullish - Peter judged it not a risk right now, but it is the second consecutive week that org has needed a lever applied to it.
Monday, July 27, 2026
A three-day window (7/24 afternoon to 7/27) with one dominant event and one dense 1:1 block. The dominant event was the LG U+ package-support escalation, which Peter resolved on two levels at once. On the substance he gave a definitive no to supporting cross-distro DNF/repo installs, and did it by installing the criterion before the technical debate started - is this something we would ship down the middle for everybody, not is it a bug and not will it close the deal - explicitly cutting off the deal-value framing when it was offered. On the machinery he codified the operating rule in writing: engineering fixes bugs without Product involvement, answers out-of-scope feature requests with a no, and the only step after that no is Product - not another escalation. He attached three companion rules: re-escalation after an answer stops, engineering stops saying you can try it and it might work in favour of that is not supported, and CIQ presents a unified front so no internal voice tells a customer something is a bug after engineering has ruled it is not. He named the root cause as five years of saying yes to close deals and accepted that the untraining will only happen through some pain. The Max 1:1 was the weeks other centre of gravity, and the through-line was Peter approving substance while re-cutting mechanism. He said yes to naming Andrew tech lead of NARF with full merge authority despite saying plainly he does not believe in the anointed-tech-lead model - then routed the act of naming to Max specifically so the org would not learn that authority flows from a CTO blessing. He granted Max standing permission to publish public Spark benchmarks with no pre-approval, overriding Maxs own offer to write an internal review doc first, on the grounds that he was about to spend the rest of the hour complaining that NARF stalls on permission-seeking. And he made metric design his single biggest ask of Max - ten automatically reported metrics spanning Justin and Nathan by Thursday - choosing measurement as the primary lever on Linux velocity ahead of personnel action, while keeping the people conversation with Nathan framed around achieving those metrics. On Everfox desktop he set the boundary before the requirements arrive: CIQ is an integrator, not a builder of novel security-critical desktop technology, and Max with Bjorn is to convert the ambiguity into integrator-shaped requirements rather than resource its worst-case reading. Smaller calls followed the same routing instinct. He ruled Ascender-orchestrates-Warewulf into Ascender Pro as paid differentiation and sent Joseph Tates proposal to Zarina rather than absorbing it. He gated future team meetups on a written purpose and a purpose-derived attendee list, anchored against the companys deliberate remote choice. He told Brady to raise the kernel-team cloud-resource block in the open Engineering Weekly rather than handle it privately, attaching one test - does Justin agree there is a problem. And he widened the Endura proposal for CIQs own build pipelines to Fuzzball, Warewulf, Ascender and Depot while dropping Apptainer, using customer exposure rather than dependency count as the inclusion test, and doing it before pricing so the chosen scope is what gets quoted. One further decision was logged sensitive=1 and is deliberately absent from this summary.
Wins
The LG episode produced a clean, reusable test - would we ship this down the middle for everybody - and Peter applied it live in front of the people who most needed to see it applied, then wrote the general rule down the same afternoon rather than leaving it as a one-off ruling. The 6/30 Everfox gate paid off: rather than CIQ absorbing an unbounded desktop ask, Everfox narrowed and simplified its own requirements, letting Peter set the integrator boundary from a position of strength. Peter also caught the self-consistency problem in the Max 1:1 in real time - he was about to argue that NARF stalls on permission-seeking, so he refused a permission request rather than granting one. On the TPS side, Performance Intensive Computing cleared a large block of Fuzzball v4.1 work in the last two weeks and Customer Engineering closed out five items including the deep-support tooling programme that was not due until mid-August.
Challenges
The LG resolution exposed that the no-re-escalation rule has no companion route: Ryan asked directly for a shortened escalation path for sales, and the written doctrine defined the terminal node without defining how to reach it quickly. Justins point cut deeper - the real failure was the initial hedged answer that was never escalated for a decision, so the process failed at the front, not at the end. On Linux velocity, Peter is now on his second lever: the people-management doctrine has been restated for six weeks without converting, and this week he shifted to instrumenting metrics instead. NARF itself is the concrete symptom - checked in fragile, left broken, reporting garbage, and the intended habit of fixing the tool rather than working around it has not stuck despite repetition. Overdue creep continued across Core Distro with nothing resolved in three days.
Friday, July 24, 2026
Three days (7/21-7/24) dominated by two themes: closing out the Google/GDC contract endgame, and hardening the machinery by which engineering communicates commitments. On Google, Peter released the 6.18 build he had been holding as leverage - having judged the hold had produced its concession - while explicitly retaining the lever on future deliverables until Google resolves the LTS question, and separately decided to slow-roll Googles request for kernel-build documentation until the contract signs, reading that request as Google preparing to insource the build. He routed the question of whether to tell GDC the 8/8 MIP listing is at risk to Kelly Hall with a criterion rather than a verdict, reclassifying a disclosure question into a contract-timing one. On the machinery side he narrowed the NGD board to cross-company coordination only - not a deliverables tracker, not a complete decomposition of product acceptance criteria - and codified the same date-plus-confidence discipline for Karl Lowenbjers team, while requiring that any agent or LLM touching a system of record do so through inspectable static code rather than directly. He instituted a rotating per-team deep dive at the Tuesday staff meeting anchored on empirical metrics whose definitions get argued out in front of peers, handed mainline TPS ownership to Ryan pointed at the NGD board, and dropped his own private fork of that report. On people: he granted Ryan one US headcount to build forward-deployed engineering in-house for Citadel, backing Bjorns strategy while refusing Bjorns capacity assertion; he brought Ryan to SKO by making the ask himself; and he blocked the AI-cost workstream on complete per-individual measurement so each persons correct plan can be determined, while approving a single-seat competing-model CVE trial. He cleared Greg to move on the open-source AI-training-data initiative with no gate, and gated an Endura pass-through offering on it consuming zero engineering this half. Three days (7/21-7/24) dominated by two themes: closing out the Google/GDC contract endgame, and hardening the machinery by which engineering communicates commitments. On Google, Peter released the 6.18 build he had been holding as leverage - having judged the hold had produced its concession - while explicitly retaining the lever on future deliverables until Google resolves the LTS question, and separately decided to slow-roll Googles request for kernel-build documentation until the contract signs, reading that request as Google preparing to insource the build. He routed the question of whether to tell GDC the 8/8 MIP listing is at risk to Kelly Hall with a criterion rather than a verdict, reclassifying a disclosure question into a contract-timing one. On the machinery side he narrowed the NGD board to cross-company coordination only - not a deliverables tracker, not a complete decomposition of product acceptance criteria - and codified the same date-plus-confidence discipline for Karl Lowenbjers team, while requiring that any agent or LLM touching a system of record do so through inspectable static code rather than directly. He instituted a rotating per-team deep dive at the Tuesday staff meeting anchored on empirical metrics whose definitions get argued out in front of peers, handed mainline TPS ownership to Ryan pointed at the NGD board, and dropped his own private fork of that report. On people: he granted Ryan one US headcount to build forward-deployed engineering in-house for Citadel, backing Bjorns strategy while refusing Bjorns capacity assertion; he brought Ryan to SKO by making the ask himself; and he blocked the AI-cost workstream on complete per-individual measurement so each persons correct plan can be determined, while approving a single-seat competing-model CVE trial. He cleared Greg to move on the open-source AI-training-data initiative with no gate, and gated an Endura pass-through offering on it consuming zero engineering this half. Three further decisions were logged sensitive=1 and are deliberately absent from this summary.
Wins
The Google hold worked and Peter recognised the moment to spend it - the build shipped, the signature stayed live, and the lever on future deliverables was kept rather than traded away. Two long-running operating-model decisions landed concretely this session: the Tuesday-staff restructure now has a named first presenter and a metrics requirement, and the leveling review is filled out org-wide and already being applied to real cases. The C3 gate held exactly as designed - Product declined to fund it, so engineering never absorbed it. Peter also publicly credited a negative result as good work, putting on the record that learning something does not work counts as a successful outcome. And he un-forked his own TPS report, deliberately exposing his reflection input to whatever the org actually maintains.
Challenges
Peter is running three simultaneous demands on the same Jira surface - the 7/21 hygiene mandate, the NGD scope narrowing, and the date-plus-confidence ladder - against a Product organisation that had been reading the board as a deliverables contract. Brady still reports he cannot tell when his top priorities will be done; Peter maintains engineering says the data is there. That gap is now framed as name-the-line-item rather than resolved. AI and hardware spend are both escalating toward board-adjacent visibility while Peter cannot see his own token cost on any dashboard, which is why he refused to let any optimization proceed on partial data. The TPS delta shows the overdue tail thickening rather than clearing: seven Core Distro items are 24 to 52 days past target, and Justins section lost its target dates entirely between the 7/21 and 7/24 reports, which removed those items from the upcoming-deadline view without resolving them.
Tuesday, July 21, 2026
Twelve-day catch-up window (prior reflection covered through 7/9 evening), dominated by the Google/GDC 6.18 relationship and a hardware push. Peter instituted a delivery-control gate on the finished 6.18 kernels - nothing ships to Google without his say-so - holding the validated image as contract leverage while deferring the final release call to Bjorn/Kelly, and convened the business side (Kelly) on the FIPS CAVP-retrigger-vs-fork question that threatens the 8/8 MIP. He approved GPU hardware spend (~60K server plus ~240K of GB200s) and drove hours-not-days urgency to stand up a Fuzzball/Arcee test cluster for board and Nvidia demos. On process, he codified the Product-Engineering operating model with Brady and Brian: value drivers are context, product owns prioritization and demands date-plus-confidence while staying out of engineering how, and engineering runs a sustainable pace but must hit intentionally re-baselined dates. He paused Steve AI Data Management policy pending Greg open-source-model direction, and backed keeping the Citadel forward-deployed/TAM function in-house under Ryan while owning the capacity assessment. One additional decision was logged sensitive=1 and is excluded from this summary.
Wins
Held the Google negotiating line with discipline - both 6.18 images built and validated by 7/17, comms centralized through Peter/Kelly, image kept as leverage rather than given away ahead of signature. Cleared budget and drove urgency to unblock the GPU cluster for high-visibility board and Nvidia demos. Landed a clear, reusable Product-Engineering operating contract that gives the PMs a clean interface (date, confidence, prioritization). Derived outcomes confirmed several prior calls are paying off: Owen Fuzzball-RC evaluation (5/5, very happy and autonomous), the Toyota feature-not-bug ruling held firm under sales pressure, and the CI/CD north star is landing via Max signed one-command spark-rocky build.
Challenges
Google/GDC keeps dragging the contract while needing the deliverable, and both the ESV cert slip and a fresh upstream 6.18 CVE landing now threaten the 8/8 FIPS MIP. Engineering is carrying a heavy overdue RLC tail (secure-boot shim, RLC Pro AI, NIST 800-171, NARF) with Nathan on PTO, at the same time product/engineering interface friction is surfacing. The recurring liability of customer-specific custom kernels (Toyota) continues to demand Peter hold the line against treating non-defects as bugs.
Thursday, July 9, 2026
Full 7/9 window (the prior reflection covered through 7/8 evening); a day of back-to-back 1:1s and H2 planning prep. Two decisions logged. The one recordable in this summary: Peter ruled the Toyota/Yoshi kernel patch a feature, not a bug fix, closing the loophole where relabeling a request as a bug lets it jump the engineering queue automatically. Support may still be offered, but priority now runs through the normal Peter+Bjorn trade-off rather than a severity-label backdoor. A second decision was logged sensitive=1 and is excluded from this summary. Peter was silent in Department Leadership and absent from the Eng/CS Standup; PIC standups and the Sovereignty white-paper session did not involve him.
Wins
Clean process-protection call on the Toyota patch: named the feature-vs-bug mislabel and shut the queue-jump path without refusing to support the customer, reinforcing it the same day when Art tried to reopen it. Greg relayed unsolicited praise that Peters one-hour help to the TileDB CTO on AI adoption was humble and genuinely inspiring to their team. On the delivery side, the CLK 6.18 FIPS-for-Google item holds at 70 percent on the critical path.
Challenges
Continued Linux delivery slippage across Core Distro under mechanical overdue-creep: NIST 800-171 now 37 days overdue (58 percent), the 2023 secure-boot shim and the RLC Pro AI 6.18 kernel each 9 days over, NARF 20 days, Kernel CI Automation 13 days, AMD Enterprise GPU turnkey 27 days and blocked. The committed Google 6.18 minimal-kernel delivery stays on the critical path amid the ongoing 0-day CVE work.
Wednesday, July 8, 2026
Short window (7/7 afternoon to 7/8), dominated by a 0-day CVE fire drill colliding with the committed 6.18 minimal-kernel delivery to Google. Three decisions logged. (1) Peter backed Nathan call to put CVE remediation first over the Google release, accepted a bounded slip (no more than a week), and personally owned a proactive, value-framed slip email to Google - correcting my initial mis-attribution that the prioritization was his call rather than a support decision. (2) A soft pass on Linux candidate Patrick Culp - no black mark, re-engage in 3-6 months - to preserve background breadth in Nathan org. (3) A conditional willingness to pull the orphaned, degraded C3 system into engineering, gated on Product actually declaring it a priority.
Wins
The 0-day CVE response executed cleanly this week - zero-days identified, patches pre-built, team mobilized, and a proactive value-framed heads-up to a customer at the signing table rather than silence. On that evidence the standing CVE-response overhaul earned a 4/5 outcome. Disciplined, consistent gating on both the hiring call and the C3 ownership question: capacity and headcount follow a clear priority signal, not reflex.
Challenges
Real-time prioritization under a three-days-to-deadline CVE collision while Google is at the signing table - capping the slip and protecting the relationship at the same time. Continued delivery slippage across Linux engineering deadlines (TPS overdue creep on Kernel CI and NARF). An unowned, degraded C3 blocking a live Huawei evaluation with no clear internal owner.
Tuesday, July 7, 2026
A short (~23h) but decision-dense Tuesday driven by an acute Linux-org staffing crunch: a second Linux departure (Roxana, a contractor, on top of Gomez) put H2 delivery at real risk. In response Peter reversed his week-old stance and removed his No on the one non-Amazon-only candidate to get a head in quickly while opening a backfill req immediately - bending the not-Amazon-blood hiring principle under delivery pressure while narrowing it to a single pragmatic exception. He reinforced (to Max and in Engineering Weekly) that an automated Linux CI/CD pipeline is one of CIQ three top H2 company objectives - a standing decision he made earlier, now re-asserted as critical to Linux delivery so it cannot be quietly deprioritized. And he firmly restructured his Tuesday staff meeting so each org owner presents their own TPS section (rather than Peter reading status to them), plus un-forking his drifted TPS report back to one shared source visible three days ahead.
Wins
Two engineering items completed in the window: the Middle East reseller partner-enablement videos (V1) and the OpsGenie-to-Jira support migration. CLK 6.18 FIPS-for-Google holds at 70 percent confidence and RESF Rocky testing-capabilities remains done. Peter also moved decisively on the Linux capacity gap the same morning it surfaced - reversing a hiring No and opening a backfill req within hours.
Challenges
Linux capacity is the headline risk: a second departure has Peter openly flagging that Linux-land puts H2 at risk from a delivery standpoint, with open reqs already unfilled. Engineering status shows continued overdue-creep, mostly mechanical: NIST 800-171 (RLC) now 35 days overdue, AMD Enterprise GPU turnkey 25 days blocked, the June-30 Core Distro cluster (secure boot shim, CLK 6.18 Pro AI, Ollama/vLLM) ~7 days past, plus carried slips on Fuzzball SaaS (56d) and RESF Mirror Manager (123d). New this cycle: fresh Wolford Fuzzball/Spark date slips entered 7/7 - DGX Spark Nvidia demo Jul 10 to Jul 24 and Fuzzball v4.1 GTM Jul 15 to Jul 24 at 0 percent confidence.
Monday, July 6, 2026
A holiday-shortened, internally focused window (6/30 afternoon through 7/6, spanning the July 4 weekend). The one non-sensitive decision of record: Peter set a hard engineering-investment gate on the Rakuten 8.6 / RT-kernel support ask. CIQ will not rebuild the torn-down 8.6 infrastructure or take on RT-kernel support (a completely different codebase) unless a signed 1.6 to 1.8M expansion funds the dedicated headcount it requires, and he armed Bjorn with the engineering objections ahead of the customer conversation. This is the same deal-must-fund-the-investment posture he applied to Everfox and the Rakuten margin line, now made concrete with a dollar floor and a two-heads rationale. Operationally, Peter personally drove an urgent Google/GDC build over the holiday to fix a 6.18 DOCA crash, mobilizing Justin, Nathan, and Skip Grube to deliver a signed build so Google could test over the weekend and walk into a Monday contract signing, while insisting that testing and validation not be skipped for one-person heroics and that the single-point-of-failure build knowledge get shadowed.
Wins
The Google/GDC DOCA crash was diagnosed and a rebuilt, signed image delivered within days over a holiday weekend, unblocking a multi-million-dollar contract signing and drawing direct praise from Google. Veeam signed and is moving to Closed Won. Engineering status posted real completions in the window: the AMD Advancing AI demo shipped, and CLK 6.18 FIPS-for-Google confidence jumped from 37 to 70 percent.
Challenges
The recurring post-deadline silent date-change pattern persists and in places worsened: NIST 800-171 (RLC) is now 34 days overdue, the AMD Enterprise GPU turnkey is blocked at 24 days overdue, and the cluster of June-30-target Core Distro items (secure boot shim, CLK 6.18 for Pro AI, Ollama/vLLM) all slipped about 6 days past deadline, alongside carried-over slips on Fuzzball SaaS (56 days) and the RESF Mirror Manager (123 days). Three Pro Hardened items sit Ready for Release but unshipped (9.7 at 100 percent confidence). Separately, the Rakuten RT-kernel ask exposed that CIQ tore down the 8.6 infrastructure when support expired, so any revival now carries real rebuild-and-hire cost.
Tuesday, June 30, 2026
A 12-day window dominated by the Doha trip (CTO Day Qatar plus MEEZA, Vodafone, Barzan, QSTP, Core42, MBZUAI - external business development in Bjorn/sales lane, not Peter decisions) bracketed by his first day back on 6/29, which produced the substantive decisions. The throughline: build accountability scaffolding under the speed-over-perfect doctrine and protect engineering from undefined scope. Peter locked a TPM charter and a one-owner-per-release mandate (TPMs solve problems and do not own engineering processes; every new process defined on a Problem/Solution/Owner grid; the incomplete NGD board owned by engineering managers and feeding the Value Drivers Board). He drew the Engineering/CS interface boundary - engineering owns defining the ask, success criteria, and documentation while CS responds to specific asks rather than digging out details. He committed to mandate Snyk vulnerability-remediation compliance across all engineering teams ahead of the fall ISO 27001 audit. On the Everfox desktop-OS deal he set the engineering-stewardship condition (a fundamentally different be-Ubuntu-while-being-RedHat business: fix hardware scope in the MSA, price the rest separately, and proceed only with explicit commitment to fund engineering regardless of revenue), tasking Max to assess feasibility with Nathan. And he is championing a Kubernetes-by-CIQ H2 engineering deliverable built on the standard while explicitly acknowledging prioritization authority sits with Product.
Wins
Returned from a successful Middle East trip (four POC handshakes for Rocky Hardened and Fuzzball) and immediately converted release-delay pain into a structural accountability fix rather than a one-off scramble. Engineering status posted real completions in the window: the RESF testing-capabilities contribution for Rocky Linux (a board-level concern) completed, the long-running SRE Unified Cloud Distribution and Depot CLI for Azure shipped, and the Pro Hardened NIST 800-171, Ansible Lockdown, and 9.7 items moved to Ready for Release.
Challenges
The recurring post-deadline silent-date-change failure mode persists and in places worsened: Fuzzball SaaS slipped 56 days (Jul 31 to Sep 25), the [BOARD] RESF Mirror Manager slipped 123 days while blocked, and RLC Pro LTS 8.10 slipped 10 days the day before its deadline. NIST 800-171 (RLC) is 28 days overdue in acceptance testing, AMD Enterprise GPU turnkey is blocked and 18 days overdue, and the Fuzzball DGX Spark demo and NARF Peridot are 11 days overdue. This is precisely the gap the new one-owner-per-release and NGD-board mandate is meant to close. The Everfox desktop-OS deal also risks opening a second front against Ubuntu that could divert the org from the core Red Hat competition.
Thursday, June 18, 2026
A dense day-and-a-half of 1:1s, the engineering Round Table, and Slack yielded six logged decisions. The throughline was margin and candor discipline under deal pressure paired with a deliberate bias toward speed over control. Peter set the Veeam escalation strategy - tell the honest intentional-tradeoff CVE story and diagnose via Dickerson before the 5 AM Monday call - and confirmed a generalized willingness to walk from bad-economics deals, currently live on Rakuten (no spending 2M to win 400K). On AI and information control he committed to the opposite of caution: reject centralized AI governance and access-guardrails on internal tools, accepting eventual personnel-info leakage as the price of adoption speed and radical leadership transparency. He scoped the EU Cyber Resilience Act down to two legal definitions and routed it to counsel before any engineering work, directed Nathan to hire for background diversity beyond Amazon, and moved to repair the mis-framed engineering all-hands with a PIC/Fuzzball follow-up.
Wins
Converted a deal-threatening Veeam CVE escalation into a clear honest-narrative plan rather than a scramble. Engineering status ticked more positive: the RLC 2023 secure-boot shim unblocked (BLOCKED to IN PROGRESS) and the long-overdue Publish Fuzzball documentation shipped. Caught and moved fast to repair a morale misfire with the PIC team.
Challenges
The Veeam escalation exposed that low-7 CVEs went unaddressed long enough for a customer to catch CIQ, with a deal on the line and a 5 AM Monday call to defend it. Chronic overdue items keep aging: the NIST 800-171 trio is now 16 days past due in acceptance testing, Pro Hardened 9.7 sits at 100 percent confidence yet 7 days unshipped, AMD Enterprise GPU turnkey is blocked and 6 days overdue, and the SRE Unified Cloud Azure Depot CLI is 2 days overdue after an 18-day silent slip. Peter also heads to the Middle East for a week, leaving Nathan in charge during a tense stretch.
Wednesday, June 17, 2026
A short window since yesterday board-meeting reflection, dominated by a customer outage and two Peter-recorded leadership meetings. Two decisions were logged. First, after the Fyr Fuzzball outage ran roughly five days before reaching leadership (the customer ultimately forced the escalation), Peter mandated a structural fix: any Sev-1 or customer outage open past 24 hours must auto-notify estaff and the account AE through tooling with no manual judgment call, and AEs get CCd at ticket creation going forward. Ryan owns building it. Second, in the C-Suite Sync Peter set an explicit north star of loosening CIQ dependence on the Red Hat kernel over time, and committed to investing now in the RESF and CIQ pipeline and tooling convergence as the infrastructure that makes that future reachable, while deliberately changing nothing for customers today.
Wins
The Fyr outage was resolved, and rather than treat it as a one-off, Peter converted it into a permanent automated escalation guardrail. Engineering status turned a corner after weeks of overdue-creep: the SRE EKS Cluster review that had been the single worst chronic blocker (long overdue and blocked) completed, the GCP deprecations slip from yesterday shipped, and the Toyota Systems x DDN partnership delivery, RESF upstream Rocky documentation, and unified support tooling all completed.
Challenges
The Fyr incident exposed that a multi-day Sev-1 reached leadership only because the customer escalated directly. A handful of items keep aging: the NIST 800-171 trio is now 15 days overdue in acceptance testing, Pro Hardened 9.7 sits at 100 percent confidence yet 6 days past deadline and unshipped, and the AMD Enterprise GPU turnkey work is blocked and overdue. The recurring post-deadline date-change failure mode reappeared on the SRE side, where the Unified Cloud Azure Depot CLI slipped 18 days without a confidence update and is now a day overdue.
Tuesday, June 16, 2026
A board-meeting day capping a roughly four-day window since the 6/12 reflection. The shareable logged decision was a fix to the mid-year performance-review acknowledgment flow: when an employee flagged that Rippling forces clicking confirm receipt before the packet is actually delivered, Peter sided with the concern and split the response by ownership — an immediate in-lane fix (managers hand-deliver packets so people physically have them before acknowledging) plus advocacy to HR to change the button wording globally, with the final wording call deferred to the C-Suite. The week also landed a major engineering milestone: the Fuzzball v4.0 suite shipped (v4 Web Interface, Orchestrate on Azure, direct object management, hostpath/external volumes) — the item that was overdue-but-about-to-land at the last reflection — and Peter recognized the team publicly for hitting a tight Friday deadline. Two additional decisions were logged sensitive and are excluded from this summary.
Wins
The Fuzzball v4.0 suite shipped on a tight Friday deadline — the release that was overdue but about-to-land at the last reflection is now complete, with early financial impact as the story goes to market; Nissan RLC Pro on Oracle Cloud also completed. Peter recognized the team publicly in announcements. On the review-process issue, he backed a legitimate employee fairness concern over a convenient existing process and moved immediately on the part he controlled rather than waiting on a system change he could not guarantee in time.
Challenges
Engineering status is dominated by overdue-creep rather than new progress — exactly the visibility gap heading into the board meeting. New post-deadline date slips landed: the RLC secure-boot shim moved Jun 15 to Jun 30 on the deadline day (still blocked at 70%) and GCP deprecations moved Jun 05 to Jun 19. Chronic blockers worsened: SRE EKS Review is now 77 days overdue and blocked; the NIST 800-171 trio (RLC 58 percent / Pro Hardened 75 percent / Ansible Lockdown 67 percent) is ~14 days overdue in acceptance testing; Pro Hardened 9.7 sits at 100 percent confidence yet 5 days past deadline and still unshipped. CLK 6.18 FIPS for Google remains low-confidence at 37 percent. The recurring post-deadline date-change failure mode persists broadly.
Friday, June 12, 2026
Dense internal week after the 6/8 LA C-Suite session — back-to-back 1:1s, a 10.2 release fire drill, and performance-review work. Five decisions logged. D1 (88%, high): after a Citadel-driven escalation (~$200k contract + expansion), Peter made RLC 10.2 priority-one for Linux eng above RLC AI and H work and posted an actionable date range (inner 6/19, outer 6/30) to department-heads — but the headline call was endorsing that engineering correctly deprioritized 10.2 per the JPD board; he framed the episode as a communication failure (Bjorn got a single confusing date without the range/assumptions), not an execution failure, and was glad the team did not go rogue. D2 (80%, medium): directional intent to move Ascender (and Ascender Pro; Ledger Pro pending a Bjorn confirm) under Zorina with dedicated headcount she will hire — for now it stays under Justin, not Nathan — with Larry and Jimmy moving to sales support. D4 (76%, medium): gate any NVIDIA/Spark announcement on engineering supportability — keep it eng-only and unannounced until CIQ has something it would ship and support, then hand to Bjorn/GTM. D5 (78%, medium): re-institute and enforce lightweight monthly performance check-ins as the real accountability cadence (Peter restarting them with his own directs too).
Wins
Resolved the 10.2 fire cleanly without breaking process — validated that the JPD deprioritization held and reframed the real gap as communication discipline (give Bjorn a range, one topic at a time). Brought clean ownership to Ascender after months of ambiguity by setting the intent to move it to Zorina with dedicated headcount. Held the NVIDIA/Spark opportunity to a supportability gate despite it being live and exciting, protecting credibility with a top-tier partner. Re-instituted a monthly performance cadence and committed to model it himself.
Challenges
TPS delta is dominated by continued overdue-creep, not new progress. The recurring post-deadline date-change failure mode persists broadly (CLK Pro AI, NARF Peridot, secure-boot shim, Kernel CI Milestone with a 37-day slip, OVAL files, Pro Hardened 9.7). Chronic blockers worsened: SRE EKS Review now 73 days overdue and blocked; the NIST 800-171 trio (RLC 58% / Pro Hardened 75% / Ansible Lockdown 67%) ~10 days overdue in acceptance testing. The Fuzzball v4.0 suite is ~7 days past its 6/5 deadlines (though at 80-98% confidence and landing this week per Wolford). The 10.2 episode itself pulled Bjorn into a confusing date conversation that Peter had to personally untangle.
Monday, June 8, 2026
Quiet window (6/5 afternoon through Monday 6/8, including the weekend and Peters travel day to the LA Product Session). Two decisions logged, both from Slack DMs — no sent email, and all of Mondays Fathom meetings were sales/customer calls not involving Peter. D1 (66% confirmed): Peter killed the follow-on Icicle patents and will finalize only the single Icicle filing already in flight, driven by Nathan/teams run-with-what-you-have signal and a closing counsel filing window — the operational next step of the 5/29 Icicle wind-down (preserve the validated patent, dont expand the family). D2 (70% confirmed, medium): after Ryan flagged that a leadership packet-share had made a Peter/Bjorn performance-review split visible to the org, Peter established that a 3 is the calibration norm (not a 5) and closed alignment with Bjorn and Greg same-day — reconciling divergent standards at the top rather than tolerating parallel norms by reporting line.
Wins
Closed a C-suite performance-calibration misalignment same-day before it could harden during an active review cycle, getting CEO and President buy-in on a single standard. Narrowed Icicle IP investment to the one validated patent (3pct constrained-HPC / 11pct mixed-workload gains) and let go of speculative follow-on filings, keeping engineering attention on productization. Completed a long-standing todo (talk to Bjorn about performance management and values) — it was satisfied by the alignment Peter closed.
Challenges
TPS delta is overdue-creep rather than new progress. A batch of Jun 5 deadlines passed and are now ~3 days overdue (the Fuzzball v4.0 suite and GCP deprecations). Chronic items worsened: SentinelOne EDR deployment still at 0% confidence and now 8 days overdue (security-critical, no movement), SRE EKS Review 69 days overdue and blocked, Unified Cloud Distribution/Depot CLI for Azure 10 days overdue at 95%, and the NIST 800-171 trio now 6 days overdue in acceptance testing. Both of Peters recent structural bets (the 6/5 pipeline-convergence mandate and the Value Drivers Board work) target exactly this post-deadline date-change pattern.
Friday, June 5, 2026
Short window (6/3 evening through 6/5) following the 6/3 reflection, and 1:1-heavy rather than decision-dense; the in-person 6/3 engineering F2F was not captured in Fathom, so that days coverage stays partial. Three decisions logged. Headline (D1, 88% confirmed on first ask): Peter laid down a mandate that CIQs Linux build/test pipeline will become functionally identical to the RESFs over time — one consolidated project rather than parallel tooling, with Nathan driving and accountable for closing the CIQ-to-RESF gaps (hardware parity, cut over to Koji, mirrored build infra, a full validation framework) and Justins build world signed on. Peter explicitly chose coordination over speed even though it slows Ryans faster build-it-now instinct; the driver is a healthy upstream so CIQ can patch zero-day CVEs for Rocky the same day. D4 (82% confirmed): Peter granted Chris Wolford standing latitude to recruit continuously (always be fishing, no open req required) — a trust-gated operating model to raise team pace and decouple recruiting from the slow req-approval cycle. One further decision this window was logged sensitive.
Wins
Landed the pipeline-convergence mandate as a structural fix for the recurring post-deadline date-change failure mode that is visible again in todays TPS (Pro Hardened 9.7 rebaselined May 18 to Jun 11 on Jun 4, 24 days post-deadline). Held coordination-over-speed against the pull to move fast and silo, and put accountability with the single owner whose adoption makes the platform real (Nathan) rather than the faster-moving party.
Challenges
A light, two-day, 1:1-heavy window rather than a decision-dense one, and the most substantive in-person source (the 6/3 engineering F2F) was not captured in Fathom, so the records coverage of that day is necessarily partial.
Wednesday, June 3, 2026
Short window (6/01 evening through 6/03) following the 6/01 reflection. Two decisions logged, both Peter-attributable; the rest of the windows meeting activity belonged to Bjorn or Greg. D1 (e250f8a4, 88% confirmed, distinct decision per Peter): in a recorded working session during the in-person engineering F2F (Peter + Nathan + Justin + Chris Baek), Peter locked the architecture separating the JPD from the Value Drivers Board and decided engineering target dates and confidence live on the Value Drivers Board engineering lane as the source of truth, kept current on a tight (Friday) cadence, while JPD dates and confidence become computed properties pulled from the board rather than manually entered. This is the concrete design realization of the 6/01 commitment to restructure the Value Drivers Board, and a direct structural attack on the recurring post-deadline date-change failure mode in the TPS report. D2 (248b8086, 65% confirmed): in the HR Weekly, Greg offered Peter the performance-engineer rec; Peter chose to leave it under Greg/research for now (roughly the first four months) with shared agreement it likely transitions to engineering later, optimizing for where the hire produces value soonest rather than org tidiness. Coverage caveat: the afternoon F2F sessions (the most decision-dense part of the day) were in-person and never synced into Fathom, so this windows coverage is partial by Peters own read.
Wins
Landed the concrete architecture for the Value Drivers Board: engineering owns its dates and confidence as the system of record and JPD dates become computed, which is the structural fix for the stale-dates problem rather than policing each late update. Held the JPD-versus-board distinction firmly against repeated pressure to collapse the two into one tool, keeping each instrument tied to its own purpose (prioritization versus dependency coordination).
Challenges
A light, coordination-heavy window rather than a decision-dense one, and the most substantive source (the afternoon engineering F2F) was in-person and not captured in Fathom, so the days decision record is necessarily partial.
Monday, June 1, 2026
Short-window reflection covering ~2.5 days since the 5/29 reflection; the weekend was quiet and essentially all activity was Monday 6/01. One decision logged. D1 (912340f0, 69% confirmed) Peter committed in the Leadership Roundtable to bring a formal proposal to restructure the Value Drivers Board, explicitly sequenced as the next coordination lever now that the product-priorities board (the JPD-doctrine work closed 5/29) is aligned where he wanted it. Triggered by Lindsay surfacing marketing/product misalignment (premature RLC+AMD announcement before AMD validation; Fuzzball multicloud date churn). This is the leadership-group surfacing of work Peter already committed to in his 5/28 Nathan 1:1; existing todo 6a4ed436 (edit value drivers dashboard, Nathan input pass Wednesday) was linked to the decision rather than duplicated. No new todos created — existing todos cover execution.
Wins
Sequenced the coordination-discipline arc cleanly: having locked WHAT gets prioritized (JPD doctrine), Peter is now turning to the cross-functional visibility instrument (Value Drivers Board) — fixing the instrument rather than policing each product/marketing misfire one at a time. Committed to arrive with a concrete proposal on a days-timeline so he shapes the board structure rather than ceding it. Tied the restructure to the live trigger (Lindsay marketing/product misalignment) and to Nathan input already in flight.
Challenges
A light coordination day rather than a decision-dense one — most of Monday was triage, status-reporting, and routing rather than fresh decisions. Candidate discovery again surfaced borderline items that on inspection were continuation/triage rather than fresh decisions, reinforcing the need to filter continuation-of-prior-intent at the discovery step.
Friday, May 29, 2026
Short-window reflection covering the ~29h since the 5/28 reflection. Three decisions logged in prose-eligible form. D1 (5341d0aa, 82% confirmed) Product owns prioritization of everything in the JPD — delivered in the Brian/Brady weekly sync (Fathom 150779031) as the third and closing beat of a multi-week JPD-doctrine arc started in the Nathan DM thread on 5/27 and continued in the Baek 1:1 and RLC standup on 5/28. Today extended the doctrine to a new dimension: engineering has no protected sustaining/KTLO baseline. If product wants sustaining work prioritized, it takes a numbered slot on the board, and anything not listed should be assumed to receive zero engineering attention. Peter explicitly named a posture shift from Socratic to directive in-meeting, used the Apple-pre-iPhone 18-months-of-zero-Mac-engineering exemplar to prove the sustaining-baseline-is-a-choice point, and accepted future Bridge quality bleed as a feature rather than a failure. D5 (4c582195, 65% confirmed) Icicle wind-down — Peter formally took Icicle off his active tracking list in the Nathan 1:1 after Nathans independent revenue-path assessment (nobody is going to pay for this). Patent prosecution continues, Fuzzball-side opportunistic feature framing preserved, but Peter attention slot freed. Same doctrine applied to Peter own focus list. D6 (cb738912, 62% confirmed) Rippling Jira Asset Mgmt add-on approved via 5/29 Slack DM with Steve Wallace, contingent on Steve securing a 12-month commit with cancellation for convenience at month 11+ rather than the 19-month lock-in originally proposed. Build-vs-buy framing explicitly named at current company size; reassess at 2x size. One additional decision was logged sensitive=1 and is intentionally not described in this summary.
Wins
Closed the JPD doctrine arc with Brady/Brian in the weekly sync — the third and most explicit beat after the Nathan DM thread (5/27) and the Baek 1:1 + RLC standup (5/28). Same-week three-surface signature matches the doctrine-lock-in pattern: same message, multiple audiences, same week. Posture shift named explicitly in-meeting (Socratic to directive) rather than left implicit. Applied the same engineering-attention-as-finite-resource doctrine to Peter own weekly-review surface via the Icicle wind-down — doctrine and personal practice aligned. Honored the build-vs-buy inflection-point framing on Rippling with the negotiation lever (cancellation clause) preserving optionality at low cost.
Challenges
Two candidate-discovery corrections this session: one candidate was anticipated-execution-of-prior-intent (not a fresh decision); one was process-improvement feedback on reflection-summary propagation rules. Both are curation calls rather than factual errors, but signal candidate-discovery is over-eager — surfacing items that are continuation/execution rather than fresh decisions.
Thursday, May 28, 2026
Short-window reflection covering 5/26 evening through 5/28 morning. One decision logged after walking through four candidates with Peter: D1 (80% confirmed) JPD board scope locked to ordered-priority-list-only — Peter delivered the doctrine to Nathan in a ~25-message Slack DM thread on 5/27 morning, then reinforced it same-day in the Baek 1:1 and at the RLC cross-functional standup where Nathan was tasked with drafting the formal product-board structure proposal. The absolute language (NOTHING more, NOT project planning, and for no other purpose) signals genuine intent rather than rhetorical emphasis. This sits in the same governance-redesign arc as confidence-as-contract (5/21), no-bugs-only-change-requests (5/19), and engineering-owns-QA (5/21) — the product/engineering interaction layer is being re-drawn one piece at a time, and JPD scope is today's piece. Three candidates were not logged: D2 Arcee priority shift was Bjorn's decision (Peter supports but did not make it; team-structure factual correction: Fuzzball and SaaS are the same team, focus is shifting to Spark work), D3 was skipped at Peter's direction, and one additional decision was logged sensitive=1 and is intentionally not described in this summary.
Wins
Same-day reinforcement of the JPD-scope doctrine across three surfaces — Nathan DM (didactic), Baek 1:1 (recap with adjacent stakeholder), RLC standup (Nathan tasked with proposal). The absolute language and the explicit push-back-when-anyone-tries-to-widen-scope instruction make the doctrine durable against the natural drift toward using JPD as a project tracker. Peter locked scope BEFORE Nathan drafts the formal product-board proposal, which is the right ordering — redlining a doc is more expensive than redirecting the author. Also: 14-minute Gauntlet CVE validation breakthrough from Ryan's org (12 kernels validated wall-clock; multi-QA-engineer-days work compressed) showcased the engineering-owns-QA + Ryan's-team-builds-tooling-not-rescue doctrine landing exactly as designed.
Challenges
Two attribution corrections in the same session — a person_id error misattributing Bjorn's Arcee priority decision to Peter, and a factual error on the team structure (Fuzzball and SaaS framed as separate teams when they are the same). The prior lesson family (calendar-attendee ≠ decider; product priorities default to Bjorn) was loaded but applied too weakly to a meeting where Peter was actively engaged. Verifying CTO-vs-Head-of-Product attribution needs to be a first-pass filter, not a second-pass check.
Tuesday, May 26, 2026
5-day reflection window (5/21 → 5/26) anchored by a single Thursday (5/21) of 1:1s with all six direct reports plus Sarah, the Friday (5/22) Engineering QBR (3 hours, full leadership), the Greg-driven Sherif/RESF escalation (5/23), and Memorial Day weekend boundary-setting (5/24). Thirteen distinct decisions land, with two clear structural themes: (1) extending and applying the doctrines codified in the prior reflection — confidence-as-contract was reapplied to Chris W on V4 (D1), engineering-owns-QA was extended to docs content with Ryan/Lindsay owning packaging (D4), and Greg AI sandbox got firewalled from go-to-market plans (D7); (2) org-line redrawing around Justin/Nathan capacity — Ascender lands in Nathans org with possible Zarina-led sister team (D3), Kyle conditionally moves to Greg gated by Wolford filling the open rec first (D2), Cedric written off the Fuzzball team (bundled into D2), Wesley exit planned. Additionally: Wallace owns Sherif/RESF systems escalation directly after Greg pinged in #distinguished-leaders (D5), hardware lab gets a paint-the-whole-picture-and-buy-it-all directive (D8), Joseph comp ratified plus a story-shaped raise-ask process (D9), CIQs first-ever bug bounty decided for Yesh (D11), cloud-security engineer hire JD path established (D13), and on Memorial Day Sunday Peter held holiday boundaries against Mariahs review-deadline pressure AND told Ryan to stop reading reviews (D10). One correction stored: Dieter Middle East trip — Peter actually decided WITH Bjorn that the trip stays and the leadership move is off-loading other stress vectors, not cancelling (D12). Short same-day gap-window reflection covering 8:54 AM → 5:12 PM PDT after the morning reflection. Six decisions landed — four operationally and two with named-attribution corrections. The Kyle/Cedric research-team placement from yesterday refined into two distinct decisions: Cedric goes full-time to Gregs research org (firming up yesterdays written-off Fuzzball-team status), and Kyles move conditions get more protective (permanent-no-return, gated on RLC completion, Greg owns the conversation). Peter committed a 12-month engineering hiring cadence to Mariahs forecasting: 2-3 net new engineers every 6 months, Linux Security as the next priority (target 4-6 months out), Bay Area hiring preference. Engineering scope was written down twice: Peters narrow Computex condition (production-ready product, not POC; Bjorn separately owns the 6-week-lag and PR-runway GTM gates) and the same-day RHEL patching support document (32 minutes from LRT action item to circulated draft, with an explicit Ubuntu carve-out making future Ubuntu requests a new-decision event). One additional decision was logged sensitive=1 and is intentionally not described in this summary.
Wins
Engineering QBR landed strong — process wins (Product Priorities Board reducing interrupts, coordinated CVE response, SRE early involvement) were celebrated and action items were assigned with clear owners. Direct reports came prepared and Peter cleared substantial decision backlog in a single Thursday. CIQs first bug bounty decision made same-day and posture-correctly. Wallace responded to the Sherif/RESF escalation in minutes within the same Slack thread. Same-day execution stood out: 32-minute turnaround from the 3pm LRT action item to a circulated RHEL patching support Google Doc with explicit Ubuntu carve-out. Cedric placement firmed up in time for Mariahs forecasting deadline; multi-meeting Kyle/Cedric arc converged into a clean swap. Hiring cadence committed at the meeting where Mariah needed the input — 2-3 / 6 months across a 12-month window. Engineering-CTO line held narrow on Computex (production-ready only) without reaching into Bjorns GTM-cadence lane.
Challenges
Sherif/RESF systems sat for weeks unaddressed and only surfaced because Greg pinged Peter publicly — SPOF (Moody bereavement leave) exposed how brittle the support routing is. Ryan brought significant emotional weight to his 1:1 (former mentees suicide); Peter held the space well but the broader mental-health-at-leadership question stays unanswered. V4 is going to slip — the slip itself is the right call but it underscores that confidence-as-contract is still being taught. Cedric drift toward Greg created planning ambiguity that needed explicit cleanup. My initial Dieter-trip hypothesis was wrong direction — I assumed investigate-leaning-cancel when the actual decision was investigate-leaning-stay-and-care-elsewhere. Same-day pivot on Cedric (written off yesterday → placed today) is a 24-hour shift driven by new risk-asymmetry data from the C-Suite Sync. The Computex window creates real pressure on Fuzzball-on-DGX-Spark productization timing — Wesleys weekend POC must convert to production-ready inside a few weeks for any announcement to be viable. Kyles RLC commitments still need to be defined before Greg can have the conversation that triggers the move. My initial framing on Computex conditions bundled three gates under Peters name when only one (production-ready) is his — the others are Bjorns.
Thursday, May 21, 2026
Two-day window (5/19→5/21) defined by three companion structural reframes landing in parallel: (a) D3 Jira confidence-as-contract doctrine codified with Justin, applied same-day by Chris W (V4 +7d at 98%), to be rolled to Nathan/Max next; (b) D4 engineering owns all QA — Nathan named accountable, tests required for Done, Ryans team re-shaped to tooling not rescue; (c) Brian Dawsons GTM-side product priority ticket consolidation (~130 → ~50-55) running in parallel on the other side. Together: the integrated engineering↔GTM trust-restoration package. Major org/people decisions: D1 Kyle AI transfer approved with conditions + Gregs AI team codified as research firewall (Cedric assumed unavailable for Fuzzball planning); D2 Ascender consolidates under Nathan, customer-facing-delivery reorg named as future possibility (Zarina as potential lead). Two time-over-cost capacity-acquisition decisions same week: D5 lab hardware bulk-buy (8-10 servers, Texas DC) and D8 Rippling Asset Mgmt $18k-for-EOJ over $15k-for-Dec. Security posture strengthening: D6 Yesh bounty precedent + D9 dedicated cloud security engineer JD to be championed by Peter. D7 Moody Citibank card hard cutoff 6/15 as canonical Accountability Follow-Through (9 months of soft asks → past-tense + future-fact framing). D10 Dieter three-layer wellness intervention (trip cancellation offer + RESF Mattermost kudos + RESF governance reform). Tactical context: Rakuten RFP submitted (positive Suzanne feedback), 249-Q Google security questionnaire near complete, mid-year reviews due Sunday 5/24.
Wins
- D3+D4 doctrine landing same week as Brian Dawson GTM-side board cleanup is the integrated reframe Peter has been working toward — three companion fixes (engineering dates + quality + product priority) shipping together is the systemic-fix-package, not three unrelated items - Chris W applied D3 doctrine same-day Justin learned it (V4 +7d at 98% confidence) — the smoking gun that the doctrine landed cleanly without prompting - D1 codified two new doctrines while making one retention call: internal-mobility-at-external-hiring-bar AND research-team-as-bounded-firewall. Single decision, multiple structural commitments. - D2 closes the open application-engineer reporting-structure todo from last reflection AND raises the customer-facing-delivery reorg as a named possibility — tactical-close + strategic-open in one move - D5+D8 articulated time-over-cost as a transferable principle Steve can carry forward to next vendor decisions — coaching by naming the lens - D6 Yesh bounty: first-of-kind decision sets the discretionary-case-by-case-no-formal-program posture as the CIQ template - Multiple JDs in flight (Ascender app eng + cloud security eng) — proactive talent investment cadence holding through Rakuten-recovery week
Challenges
- Pattern-internal-consistency check: Peter modeling hard deadlines for Moody (D7) while his own 5/12 commitment to chase Greg on AWS Mirror Manager is STILL open per Sarah Sync 5/21 — Greg-ask todo has not moved in ~10 business days. Ironic timing. - TPS delta shows Justin pushed RLC Pro LTS 9.6 (May 15 → May 29) ON 5/21 — exactly the kind of post-deadline push the 5/19 + 5/21 reframes were meant to prevent. Doctrine has not yet landed across all directs (which is precisely why the D3 Nathan/Max walkthrough todo matters). - Chris Wolford pushed Fuzzball Orchestrate object cache (May 22 → June 5) ON 5/21 — same pattern, though Chris is at least pushing pre-deadline. - Mass 5/15 deadline slip continues to age across Core Distro + Release All Things (Errata-CEM, shim-review, BootC, RLC+ ISOs etc.) — Nathan officially re-dated to 5/29 on 5/19, four days after the deadline. - RESF/OSPO items aging without movement (RESF Testing Team scope 55 days overdue, Improve Images 83 days, etc.) — the RESF governance reform (D10) is the structural response but its impact horizon is months not days. - Mid-year reviews due Sunday 5/24 — Phase 1 calibration gate. Rakuten + CVE recovery week ate capacity. Completion risk real. - Promised Kelly Wall the AI/infrastructure cost-category list same-day from 5/20 meeting — already late by 5/21.
Tuesday, May 19, 2026
Week was structurally dominated by Rakuten RFQ submission (Mon-Tue deadline) overlapping with continued CVE response, Peter traveling for oldest childs graduation Mon-Tue, and the Tue 5/19 board meeting today. Six decisions captured. D1: Rakuten RFQ prioritized over board prep and vulnerability work — Peter held scope-discipline at line-item level (stripped runc nohz_full, ACC100, scrubbed AI-generated CVE SLO Framework). D2: structural reframe of bug-vs-change-request terminology coached to Brady — Brian aligned, rollout via Chris/Jamie/Nathan team. D3: Values framework endorsed to Bjorn for Adam coaching conversation — efficient-and-excellent covers team-player, no-stacking-bodies as the analytical hook. D4: OPA precedent named explicitly with Nathan — three-in-a-row water-carrier tap (Maple), not single-event hero pay (Sultan/David held). D5: Ascender Pro Dev JD scope finalized — application-engineer title $180k mid-senior, panel set, JD posts weekend/Monday. D6: Bjorn-Victoria HR-leader intro — proactive talent pipeline, no slot, Bjorn owns next step. Leadership Roundtable 5/18 captured the broader cadence: hire 5 engineers + 5 sales reps in next 6mo for $35-40M 2027 revenue target, RLCAI/RLCH parity is top priority, Stacks messaging re-establishes weekly launch cadence.
Wins
- D2 structural reframe (no bugs, only change requests) is the highest-leverage move of the week — Brady self-committed to roll out without asking permission, Brian aligned. Joins the strategy-first/tactics-second framework as cumulative meta-coaching with Brady. - Rakuten scope-discipline held even through graduation-travel out-of-pocket window. Peter explicitly approved final submission only after Nathans line-item changes to 19/37/10/38/23 landed. The AI-generated CVE SLO Framework caught at final-review by Nathan was scrubbed — close call, no public commitment to 48h/72h/14d SLOs. - OPA precedent (three-in-a-row, not single-event) and Adam values-framework endorsement landed in the same week — both apply the same value principle: reward consistent reliability, not heroic-but-unsustainable behavior. Bjorn coaching alignment 5/18 mirrors Nathan coaching alignment 5/14. Coordinated values-application across both directs. - Fixed Fathom MCP get_meeting 404 issue — get_transcript works as fallback. Reflect.md and self-assessment.md updated. Would have lost the 5/14 Nathan/Steve/Ryan 1:1s and 5/15 Lindsay/Brian-Brady transcripts entirely. Process improvement made mid-reflection rather than next-time.
Challenges
- Mass 5/15 deadline slippage on TPS: RLC Pro Hardened 9.6 LTS, RLC Pro AI BootC, RLC Pro LTS 9.6 (81%), RLC Pro 9 ISOs, RLC+ 9 ISOs, RLC 9 AMD GPU, Errata CEM, CIS/STIG Lockdown Playbook, NIST 800-171 RLC and RLC Pro Hardened all 4 days overdue. shim-review refresh BLOCKED 4 days overdue. Pattern is compound: copy-fail and dirty-frag CVE response consumed the team capacity that would have hit these deadlines. - AWS access for Mirror Manager still blocking Steve five business days after Peter committed in 5/12 Engineering Weekly to talk to Greg today. Greg-ask todo open, but outcome has not moved. Five days is the threshold where the commitment-to-talk-to-Greg starts to look like a slip itself. - Brady de-prio carryover from 5/8 + 5/12 reflections still not visibly closed — 8+ business days now. Pattern of three-business-days-overdue evolved into eight-business-days-overdue. Worth a check-in with Brady whether de-prio actually happened or whether the carryover signals he is stuck. - Yordan QBR-absence conversation has now waited 5 business days from when Peter flagged on 5/12 that the longer it waits the more it reads as benign neglect rather than deliberate omission. Currently in benign-neglect territory by Peters own framing. - Mid-year reviews due 5/24 — Phase 1 of three-phase cycle. With 5 business days to go and the Rakuten week eating capacity, completion risk is real.
Tuesday, May 12, 2026
Tuesday after a long Monday window. Three decisions, all execution-layer follow-throughs on the structural-lever arc. D1 converts the 5/11 Leadership Roundtable vuln-handling commitment into three operational pillars (process redesign with Chris Baek, tooling commitments to ask Greg for Claude whitelist + stand up unbridled LLM on Fuzzball, kernel philosophy review pulled forward to early June). D2 puts a named boundary and revisit trigger on a direct reports personal-health leave extension (6/5 cap with Bjorn co-consulted on the generous-leaning decision; Mariah flagged precedent in writing). D3 sets the 5/22 Engineering QBR as a collaborative working session with three named topics — the third (streamline communication and increase work visibility) is the load-bearing one because it is the same theme that drives the Jira-as-record arc. Customer/sales meetings on the calendar today (Spectro Cloud, Cadence, SSC/Ascender) were NOT Peter — Dave Dickerson and Scott Benson recorded them; Peter not in participant lists.
Wins
- Converted yesterdays LRT commitment from C-suite promise into specific Jira-ticketable work across process / tooling / strategy layers within 24 hours — the externalize-then-operationalize cadence is now visible across three consecutive days (5/8 internalize → 5/11 externalize → 5/12 operationalize) - Got an explicit boundary + revisit trigger on the Max leave extension in writing with Bjorn pre-consulted — Mariahs precedent flag now lands against a named structure rather than open-ended generosity - Used Fuzzball as the answer to two strategic problems at once: vuln-investigation tooling gap AND the AI sovereignty dogfooding story Brady/Brian have been developing - Sarah/Peter sync surfaced Yordan absence from QBR as a separate signal worth direct conversation — caught early enough to act on this week
Challenges
- Five Greg-asks now queued in the same week (Claude whitelist, AWS Mirror Manager, Citadel support duration, Joseph Tate comp from 5/11, Ascender JD coming Thu/Fri via Bjorn-scheduled Jimmy meeting) — risk of asks-in-a-bundle diluting individual signal weight - Brady de-prio carryover from Fridays reflection still not visibly closed three business days later - Yordan QBR-absence conversation has not yet happened — the longer it waits the more it reads as benign neglect rather than deliberate omission - Mid-year performance reviews due 5/24 with the cycle only ~50% complete at the midpoint — Mariah flagged at HR weekly that directs are behind
Monday, May 11, 2026
Three-day window (5/8 PM through Monday 5/11). Light calendar Monday — Greg 1:1 + Leadership Roundtable. Three decisions, all extending the structural-lever arc into new domains. D1 executes the 4/17 Joseph Tate retention commitment — formalized in an email to Greg+Mariah after verbal alignment in the 1:1. D2 externalizes the 5/8 internal build/test infra commitment to LRT — cross-functional commitment now binding Bjorn, Greg, Chris Baek, and Lindsay; Chris owns short-term process definition, Peter owns long-term infra. D3 opens the strategic question Saturday in #department-heads: what Dirty Frag means for kernel posture, upstream binding, RESF relationship, and RLC/RLK brand meaning — staged as agenda input for the mid-to-late June LA product-strategy session with Bjorn and Greg. Customer/sales meetings on Monday calendar (UMS, UND, IMMIX, NLIT-Sandia, SHI, Fyr, Wisk, Zymo) were NOT Peter — recorded by sales team, Peter not in participant lists.
Wins
- 4/17 Tate retention commitment closed with formal proposal in 24 days — inquiry → formal ask sequence completed before attrition risk converted to attrition - Saturday #department-heads message converted Dirty Frag exhaustion into agenda input for the LA session — same exhaustion-as-diagnostic playbook that drove the 5/8 internal commitment, now applied at the strategic-identity layer - LRT vuln-handling commitment paired deliberately with Chris Baek's short-term process work; Chris pulled in Lindsay/Ramesh/Melissa to expand the doc — companion alignment achieved without separate negotiation - Nathan's team receiving $2,500 OPA for Dirty Frag remediation; Greg taking lead on the announcement (Peter approved Greg's draft) — Nathan's public credit visible to engineering org without Peter needing to author the praise - Joel/Ashish intro (5/8) — personal-network connection brokered between IAG Capital and former employer; relationship-capital deposit
Challenges
- Friday's reflection named Brady de-prioritization follow-up, LGU+ contract language verification, and NVIDIA/Scott Monday AM as Monday priorities — none of those visibly happened. Monday calendar was Greg 1:1 + LRT only; the named priorities slip into Tuesday or later - D3 RLC/RLK question opens four entangled strands but the venue is mid-to-late June LA session — risk that the framing decays between now and then if not actively kept warm in Nathan/Justin 1:1s next week - Bjorn DM about Ascender JD reveals memory/handoff drift: Peter recalls a JD coming across his desk but didn't generate one — ambient knowledge gap on hiring inventory
Friday, May 8, 2026
Short window (5/7 PM through 5/8 AM, after the 5/7 morning reflection covered Nathan/Steve/Ryan 1:1s). Two decisions, both extending the structural lever set into new layers. D1 prioritize build/test infrastructure to eliminate reactive engineering interrupts — committed in Brady/Brian sync after Dirty Frag took Linux engineering offline 24 hours; same lever as paved-paths Jira (5/7) and strategy/tactics coaching (5/5), now applied to the incident-response layer. D2 reject open-ended LGU+ RHEL/OEL support commitments — same-day intervention before signing, surfaced by Justin → Nathan → Peter; third member of the GDC/Rakuten contract-discipline arc, with Peter explicitly comparing in the Nathan DM. Hard Rock Digital sync (Greg/Patrick/Scott from CIQ) and Sovereign AI Continuity Focus (Dave Dickerson recorded) were NOT attributed to Peter — calendar attendee ≠ attendance, and neither showed Peter as participant in Fathom.
Wins
- Dirty Frag response: kernel team patched and shipped overnight with clean handoff across timezones; Brady noted in #department-heads this is a big shift from a year ago — credit Nathan for geo-diverse team design that made the round-the-clock progress possible - Brady arrived at the Brian/Brady sync ASKING for a de-prioritization proposal (per next-steps), which means the 5/5 strategy-first/tactics-second coaching has landed in his framework - LGU+ contract intervention happened pre-signing rather than post — earlier in the lifecycle than GDC or Rakuten interventions; pattern is moving forward in the contract lifecycle - Used the Mythos framing in #department-heads to publicly recast incident response as a structural problem rather than a heroics problem — sets the frame before Brady proposes what gives
Challenges
- TPS deltas continue to age 1 day each: Citadel OOM 7d→8d, Kernel CI Auto M2 7d→8d, RLC Pro Hardened 9.7 10d→11d, Self-Serve Portal Bootc 13d→14d, AMD Fuzzball POC 37d→38d (still 0%) - One fresh date-change-near-deadline: Justin pushed RLC Pro 9 AMD GPU Support May 8 → May 15 the day before deadline. Not yet post-deadline like the 5/5 fires, but the pre-deadline edge is the precursor pattern - Brady still needs to come back with the actual de-prioritization list — the build/test infra commitment is announced but not yet funded; risk of drift is real - LGU+ contract resolution is in flight — Bjorn pushed back lightly (we agreed to give answers); if final language drifts back toward SR-ticket commitment, the engineering veto becomes performative
Thursday, May 7, 2026
Two-day window covering 5/6 (San Jose office day) and 5/7 (Reno QBR day-trip) — eight decisions across hiring, performance evaluation, process structure, and technical gating. The structural lever Peter set yesterday (3-tier board hierarchy + strategy-first/tactics-second framing) executed at the operational layer this window: Justin 1:1 mandated paved-paths Jira initiative + Jira-as-system-of-record (D3) — direct response to Brady CVE post-mortem. Nathan 1:1 approved Veeam-related kernel hire (D1) and set explicit AI-inference benchmark gate for Icicle (D4) — no more drifting on synthetic results. Ryan 1:1 resolved the Owen ambiguity (D2) by routing him to Maxs AI tooling for definitive evaluation; same meeting Peter committed to ask Bjorn to deliver ARR/Series-B/dilution rationale to engineering (D7) — Product owns the why. Steve 1:1 set Moody senior-or-managed expectation framed as response to his growth ask (D6) and aligned on late-August dedicated security hire (D8). #distinguished-leaders thread surfaced senior leadership candidate position (D5): open to talk, no pressure, will not give away the kingdom — paired with Mirantis-acquired and Core42 smaller-deal context. Day-trip to Reno for QBR executed cleanly per yesterdays coverage plan.
Wins
- Yesterdays 3-tier board hierarchy converted into operational-layer execution: Justin 1:1 landed paved-paths mandate AND Jira-as-record AND ready-for-eng accuracy requirements in a single meeting — same lever, deeper layer - Owen ambiguity resolved by structured experiment (D2) rather than picking a side between Bjorn/Max/Nathan/Ryan disagreement — defensible and observable - Icicle gate (D4) is sharp: real AI inference on H100, not synthetic; prevents the project from drifting through ambiguous data - Veeam hire approval (D1) lets Nathan run Sergei + Twitter/EC2 referral in parallel — kernel bench grows ahead of urgency - Moody framing (D6) preserved development path while applying senior-bar — coaching, not punishment - Reno QBR day-trip executed cleanly per yesterdays Mariah-lunch + C-suite-intro plan; no schedule failure
Challenges
- TPS deltas continue to age: Citadel OOM and Kernel CI Auto M2 now 7d overdue (was 5d on 5/5), RLC Pro Hardened 9.7 now 10d overdue (was 8d), Self-Serve Portal Bootc 13d overdue (was 11d), AMD Fuzzball POC 37d at 0% - The post-deadline date-change failure mode that hit Wolford and Wallace 5/5 (3 fresh post-deadline changes in one day) did not produce new fires today — but the existing slips are not closing either - Senior leadership candidate flow is steady-state but each one consumes leader time at the C-suite layer; cost amortization comes only if the bar holds (D5) - Owen evaluation experiment is right but takes ~3 weeks to deploy (Max returns from leave) — gap between decision and signal is real
Tuesday, May 5, 2026
Tight 21-hour window dominated by structural execution of yesterdays push. Peter and Bjorn formalized the Three-tier board hierarchy (Strategic EPICs / Value Drivers / Tactical Jira) across three meetings — C-Suite Sync, Eng Weekly Sync, Peter/Bjorn 1:1 — with Greg agreeing to disagree-and-commit once rules are documented. The personnel consequence is staged: Brady role moves to project/program management. Peter coached Brady on the 2-step framework (strategy-then-tactics) in their weekly 1:1, then enforced it live 90 minutes later in a CVE post-mortem group DM when Brady tried to skip step 1 — Your only job is prioritization. Step one is answer my question. Step two will never happen absent step one. Never. Not one time. Brady eventually conceded. Reno QBR coverage locked: Peter delivers C-suite intro Thursday morning (Bjorn arrives PM, Greg in Houston). Justin added to tomorrow Binarly meeting; debrief sequenced AFTER not before to preserve unfiltered take.
Wins
- Three-tier board hierarchy landed across three forums in one window — yesterdays push to Bjorn converted into a structural commitment with Greg disagree-and-commit pathway - Coached Brady on the strategy-first framework in 1:1, then live-enforced it in DM 90 minutes later when Brady deviated — boundary visible to Brian as well - Auto-logged 2 high-confidence decisions (D1 93%, D2 92%); D3/D4 confirmed without correction - Yesterdays PPL-realign todo (sort 8) executed today — the alignment with Bjorn happened in the 1:1 - Reno QBR coverage solved without single-point failure: Peter early Thursday, Bjorn afternoon, Greg in Houston
Challenges
- TPS deltas continued worsening: Citadel OOM and Kernel CI Auto M2 each now 5d overdue (was 4d), RLC Pro Hardened 9.7 now 8d overdue (was 7d), Self-Serve Portal Bootc 11d overdue (was 10d), AMD Fuzzball POC 35d at 0% - THREE new post-deadline date changes happened TODAY: Fuzzball Volume Provisioners and Fuzzball hostpath volumes (Chris Wolford, 6d after deadline) and Unified Cloud Distribution Azure (Steve Wallace, 31d slipped, 5d after deadline) — exactly the failure mode Peter has been pushing on since 4/14, now with Chris W and Steve W signatures - OLAMA/VLLM ticket hygiene exposed leadership info gap because Nate is out — single-person dependency on the Ready-for-Engineering transition - Brady CVE post-mortem in DM showed how much resistance the strategy-first frame still draws from a tactical-by-default Product muscle - $5k/$15k Rippling decision drift: Wallace acknowledged the slow surfacing pattern but the meta-process fix is parked for Friday rather than decided today
Monday, May 4, 2026
Three-day window dominated by structural fixes to the Product-Engineering interface. Peter formalized three companion decisions that all share the same shape — externalize friction back to the requester so layer separation holds: (1) docs delivered via Jira exit criteria, (2) Engineering veto required on custom deals, (3) PPL must return to strategic epics with granular tasks pushed to JIRA. Two large-customer commercial calls held the line on the same principle: Rakuten gets a three-tier proposal where 8.6 support carries an $800k-$1M PS price tag, and Everfox gets a $2M front-loaded year-one floor (rejecting the $600k back-loaded structure). HR weekly opened a broader performance review window across direct reports; Mariah was authorized to escalate Stephen Moody delays directly to Peter, bypassing Wallace as a near-term workaround. CVE response continued: 9 of 10 Copy.Fail patches landed; Peter pushed back on Greg framing about the patches and committed to driving the Dieter conversation about upstreaming the work to RESF.
Wins
- Structural fixes landed as a coherent set: docs process + Engineering veto + PPL realignment all share the same layer-separation lever — three companion decisions in one window - Held the Everfox $2M front-loaded floor without softening, applying the 1/29 precedent that worked - Three-tier Rakuten proposal makes 8.10 migration the easy path and prices 8.6 support at real cost — preserves CIQ kernel team capacity - Auto-logged 2 high-confidence decisions (D2 92%, D4 93%); zero corrections through the walkthrough - Caught a 6-week Moody/Rippling pattern that Wallace had not escalated; surfaced into Peter visibility chain before the broader Wallace conversation
Challenges
- TPS deltas worsening across the week: CLK 6.18 slipped 16d (Apr 29 → May 15) AND date changed AFTER deadline by 5d, RLC Pro Hardened 9.7 now 7d overdue (was 4d), Citadel OOM 4d overdue (was 1d), Self-Serve Portal Bootc 10d overdue (was 7d), Kernel CI Automation Milestone 2 now 4d overdue - Pattern of date-changes happening AFTER deadlines (CLK 6.18, RLC+ 9 ISOs, RLC Pro 9 ISOs, RLC Pro Hardened 9.7, RLC Pro 9 AMD GPU) — exactly the date-hygiene failure mode Peter has been pushing on since 4/14 - AMD Fuzzball POC now 34 days overdue at 0% confidence — needs explicit kill-or-revive call - Ryan-Nathan Icicle testing wires-crossed: each thought the other had it, testing got de-prioritized, Ani patent window 1-2 weeks - May launch cadence (RLCAI 5/14) at risk because Ollama/VLLM tickets still in Ready for Engineering — forcing Peter to commit to firm dates by 5/5 11am
Friday, May 1, 2026
Two-day window dominated by an all-hands Linux kernel CVE response: CIQ found and is shipping an extra fix that RH/CentOS missed, contributed it upstream (acknowledged for inclusion), and pushed to be first EL distro to release. Customer reassurance is the primary goal; public proof point that CIQ contributes to security and is large enough to serve big customers is secondary. Coreweave deal closed in parallel ($10M ARR threshold crossed). Senior IC return-from-leave conversation framed by Peter as PTO + path-forward, not corrective communication; Peter held that line firmly with HR. Ryan's scope expanded materially: docs.ciq.com ownership + AI-driven Veeam image builder POC. Justin instructed to enforce Jira-as-system-of-record by refusing Slack-only requests. Mariah offered explicit peer-tier coaching with a Reno Thursday in-person follow-up.
Wins
- Found a kernel patch RH missed and gave it back upstream (acknowledged for merge) — credibility move that lands in the same news cycle as Coreweave closing - Coreweave booked, $10M ARR threshold crossed for the first time - Held the line with HR on senior IC return-from-leave framing — protected against the worst-case (no return) at the cost of mild discomfort with HR - Ryan's portfolio meaningfully expanded: docs.ciq.com + AI image builder POC + Intelligence Hub repurposed as data lake - Mariah received explicit peer-tier coaching offer + in-person Reno follow-up — used a moment of friction as a coaching window
Challenges
- Three TPS items overdue beyond prior reflection: RLC Pro Hardened 9.7 Image Release now 4 days overdue (was 2), Self-Serve Portal Bootc OCI 7 days overdue (was 5), Citadel OOM newly overdue 1 day, CLK 6.18 newly overdue 2 days, Kernel CI Automation Milestone 2 newly overdue 1 day - New 29-day slip on GCE ARM Mellanox CX-7 RDMA (Apr 30 → May 29, Nathan), and 29-day slip on RLC Pro AI OCI Container Images (Apr 30 → May 29, Justin) - RESF infrastructure project blocked 1.5 months on AWS access; pre-provisioned account bleeding ~$10k/month while idle; threatens May Rocky point release stability - RESF Board Resolution items still ~33 days overdue - LinkedIn FUD claiming CIQ behind on the CVE — public messaging response still pending Lindsay/Nathan coordination
Wednesday, April 29, 2026
Two-day window centered on Day 1 of the IAG Silicon Valley Summit. Linux engineering bench-building continued — activated the dormant Serge Hallyn introduction by handing the relationship to Nathan. While at IAG Peter assigned Justin (not Nathan) ownership of the Binarly engineering relationship, empowered Nathan to defer the Hassan/Google secure-boot working session if Engineering was not ready, and sent IAG/Sam a structured operational due-diligence note on Raven Resonance with explicit scope bounds.
Wins
- Linux engineering hiring pipeline continues unaffected — Serge Hallyn intro activated and routed to Nathan - Nathan empowered to make a real call (defer or proceed) on the Hassan working session — leadership growth signal - IAG operational note on Raven Resonance lands with disciplined scope (operational mechanics yes, market opinion no) — preserves credibility for future LP assessments
Challenges
- Three TPS items still overdue from prior reflection: RLC Pro Hardened 9.7 Image Release (74%, now 2 days overdue), Self-Serve Portal Bootc OCI Acceptance Testing (100%, 5 days overdue), Fuzzball on Oracle Cloud (95%, target slipped to May 1) - RESF Board Resolution items now 33+ days overdue (was 31 last reflection — getting worse, not better) - RLC Segmentation: RLC Pro 9 Cloud Marketplace remains BLOCKED at 96%, Chris Baek slip now 20 days - New concern surfacing: FIPS 140-3 for CIQ Linux Kernel 6.18 at only 37% confidence for Jul 31 target
Monday, April 27, 2026
Friday, April 17, 2026
April 17 was a process-discipline and positioning day. Peter repositioned Core42 from a Fuzzball-only sale to a full-stack compliance partnership — using the Tech Dive Part 2 scope revelation as a positioning lever and applying his reshape ambiguous customer asks into defensible structure play (same move used with Tissa a day earlier). He reinforced a broader quality mandate with Brady and Brian: quality initiatives must be ticketed as visible work competing for resources, grounded in the Trexel FIPS/disk encryption failure. Structural people moves: opened hiring for a dedicated Ascender engineer to fix the Quantic/Jimmy-Larry PR rejection pattern at the source, and decided to manage Wesley via the Greg channel with a Chris boat-anchor trigger rather than putting him on active performance management. GLibC/Jeremy Allison escalation root-caused as Product-prioritization bypass. Nathan shared the Core42 BIS Export License compliance matrix DRAFT later the same day.
Wins
- Fuzzball [Engineering] Observability shipped today (was overdue at 80% on Apr 16) - RLC Pro AI OCI Container Images moved to Ready for Release at 91% confidence - Core42 opportunity expanded from Fuzzball to full-stack partnership — positioning momentum - Chris Wolford committed to two weeks of perfect Jira hygiene to validate Product's post-ship change metrics (continuation of Apr 14-15 Jira hygiene arc) - Peter reshaped Wesley conversation from all-perf-issues-are-PIPs to trigger-based escalation, preserving Chris's bandwidth
Challenges
- Release CLK 6.18 now overdue by 1 day (was on target Apr 17) - RLC+ 9 ISOs and Netboot overdue by 1 day - Self-Serve Portal P4 Centralized License Management overdue by 1 day (still Acceptance Testing at 91%) - RLC Segmentation RLC Pro 9 Cloud Marketplace now 8 days overdue (was 6 days on Apr 16) — BLOCKED status continuing - Trexel FIPS/disk encryption failure surfaces that the only confirmed automated check is does it boot — CI/CD gap is broader than one customer - GLibC/Jeremy Allison work bypassed Product prioritization entirely — process failure requiring Brady to split tickets
Thursday, April 16, 2026
Wednesday, April 15, 2026
Tuesday, April 14, 2026
April 11-14 covered board meeting prep, partner alignment, retention, and accountability. Peter mediated between Greg and Max to get the Dieter/RESF justification ready for the board meeting, translating between Max's technical perspective and Greg's political needs. He proactively ran a skip-level retention check-in with Alex de Wergifosse after Ian's departure from Fuzzball, establishing a mutual transparency pact and recurring 3-4 month cadence. In Leadership Roundtable, he publicly committed to a tough date hygiene conversation with directs after ongoing estimation failures (58% miss rate), requesting magnitude data from Chris Baek. He escalated a Google/NVIDIA messaging discrepancy to Bjorn for same-day resolution — Google giving NVIDIA conflicting info about Rocky usage while engineering work is on track. He also delayed an HR action pending his own 1:1 assessment with Chris.
Wins
- Successfully mediated Greg/Max/Dieter justification chain for board prep - Proactive retention check-in with Alex surfaced positive feedback about Justin's management - Google engineering milestones confirmed on track in CIQ 6.18 meeting - TJ Gohl's AWS Identity Center self-service access management deployed — recognized by Peter and Steve Wallace - Performance review framework docs submitted to Greg via Mariah - $100k Ascender deal in procurement (created Apr 1, expected to close this month)
Challenges
- Google/NVIDIA messaging discrepancy undermining CIQ credibility with partners - Engineering date slips continue — CLK 6.18 overdue 4 days, multiple items with 28-30 day slips - Steve Wallace pushed HR Onboarding 49 days and RESF Mirror Manager 31 days - Multiple OSPO items 18 days overdue - Bjorn's flight delayed for board meeting today — may need to cover opening
Friday, April 10, 2026
April 10 was a heavy strategic alignment day. Peter defended the global product prioritization model to Greg (post-Exec meeting where Fuzzball was elevated to top 20), resolving Greg's concern with a per-team computed property. He set a critical Lab-to-Production boundary with Bjorn — nothing from Cedric's lab goes to production without Engineering productization. In the Brian/Brady sync, Peter reinforced that Product owns exit criteria (triggered by Engineering unilaterally removing NVIDIA CUDA from RLC Pro 9.6 LTS), surfaced a 58% estimation miss rate, and committed to attending Engineering Weekly for accountability. He then forwarded that meeting recording to Bjorn and Baek for alignment on the prod/eng interface. In #distinguished-leaders, Peter shaped the performance engineering hire interview panel, retaining veto power for himself and Max when Bjorn was dropped to make room for Cedric. The day closed with a Value Drivers Board Review where Peter, Bjorn, and Baek agreed to an 'Uber Value Drivers' two-tier framework for strategic clarity.
Wins
- Resolved Greg's Fuzzball prioritization concern with the global+computed-property framework — Greg confirmed: 'I feel much better now' - Aligned with Bjorn on Lab-to-Production boundary before the formal handoff meeting - Clear prod/eng interface directive delivered and shared with leadership (recording to Bjorn + Baek) - 3 PIC items now Ready for Release (NextSilicon Maverick-2, fuzzball-cli cloud, Consolidated Fuzzball interfaces) - Kernel CI Automation Milestone 1.5 completed; FIPS 140-3 moved to In Progress
Challenges
- 58% engineering estimation miss rate — most changes occurring after target dates - Multiple TPS items with date slips continuing (secure boot key blocked, CLK 6.18 due today at 82%) - Lab/Cedric delivery gap — Greg's vision for 'performant' space still vague - Need Greg's buy-in on the Lab-to-Engineering handoff boundary (not yet secured)
Thursday, April 9, 2026
April 8 was a high-intensity partnership day — Google GDC follow-up and NVIDIA EBC at NVIDIA HQ back-to-back. Peter coached Brady and Nathan with detailed communication directives before the Google meeting, emphasizing controlled messaging, distinguishing technical infeasibility from resource constraints, and maintaining CIQ's calm/capable/dedicated image. He personally bookended the engineering meeting with success criteria. In parallel, Peter scoped the NVIDIA Grace Vera partnership, committing to Scott Hara to assess headcount needs and forwarding patch lists to Nathan (SWAG: 3-6 months RLC). On the people front, Peter initiated proactive retention check-ins after Ian's departure — Sultan confirmed widespread team fear and thought the meeting was a termination notice. Peter clarified Michael Young's departure was performance-based and reaffirmed the A-player upgrade strategy. In #distinguished-leaders, Peter and Bjorn confirmed proceeding with the RESF Dieter restructure despite Greg's hesitation about disrupting Leigh's momentum. Peter also set a value-first attendance framework for Nathan at Google NEXT — only go if there's a concrete business objective. NARF automation progressed with 9 packages shipped to production.
Wins
- NVIDIA EBC at HQ with VPs — strong executive engagement on both sides - Google FIPS 618 meeting had productive outcome — new libcrypto collaboration with Google maintainer - Tissa acting as strong internal champion, setting 2-week deadline for Madhu's scale numbers - NARF shipped 9 packages to production (Andrew driving while Max OOO) - Nathan delivered quick SWAG on NVIDIA patch assessment same day - CIS Benchmarks Barricade Combined Playbook completed
Challenges
- Widespread team fear confirmed by Sultan — the culture of fear is deeper than expected - Nathan's team stretched thin: Google FIPS 6.18, NVIDIA patch assessment, Google NEXT question all competing for same bandwidth - Multiple Fuzzball items overdue (Volumes 6d, Observability 10d, GCP integration 6d) - Still no scale numbers from Madhu at Google — Tissa escalation is the backstop - RLC Pro Hardened 9.6 LTS slipped again (now Apr 24)
Tuesday, April 7, 2026
Day dominated by the Google FIPS crisis escalation and strategic clarity push. Peter took personal control of all GDC communication for the next month after identifying that inconsistent, over-transparent messaging was weakening CIQ's negotiating position. In the Brian/Brady weekly sync, Peter called the ~50 value drivers a 'shotgun approach' and demanded consolidation to ~3 core drivers, requesting a 1-year product vision for RLCAI/RLCH from Brian. By evening, with Google's Madhu stalling on revenue projections, Peter drew the line: FIPS delivery depends on getting a ramp or a new contract. The FIPS execution continued — Tissa authorized, atsec contract signed, Nathan directed to begin Option 2 engineering work. Peter shared Anthropic's Glasswing article with leadership, sparking interest in AI security auditing of open source code (Jeremy Allison flagged potential CVE wave in 20-year-old NFS code). Ryan Smith delivered a comprehensive RESF team-by-team status update showing strong operational progress. Steve Wallace's surgery went well. Ryan Smith will be OOO Apr 11-18.
Wins
- Tissa authorized FIPS work and atsec contract signed same day — fast execution - Strategic clarity emerging: ~50 value drivers being forced to ~3 core drivers - Ryan's RESF operational update shows momentum across all teams - Steve's surgery successful - Anthropic Glasswing conversation opened potential AI security angle for CIQ
Challenges
- Google appears to be stalling on revenue projections, possibly exerting leverage - GDC communication has been inconsistent — required Peter to centralize control - ~50 value drivers reveal lack of company-level strategic clarity - Multiple TPS items slipping: CLK 6.18, Cloud Marketplace, HubSpot integration all pushed by Brady - Ryan OOO Apr 11-18 creates coverage gap during critical RESF momentum period
Monday, April 6, 2026
Sunday work day dominated by the Google GDC relationship crisis. Peter took personal ownership by drafting and sending a strategic email to Manu at Google, offering FIPS 6.18 acceleration at certification cost only (~$180k, absorbing NRE), directing a shared git repo for co-development, and requesting revenue ramp projections. This is the execution of the FIPS leverage strategy identified in the Apr 3 reflection. Simultaneously, the GCE side of Google is thriving — Hassan from GCE will pitch Rocky Linux to NVIDIA VPs Wednesday, armed with data showing explosive Rocky growth vs flat Ubuntu on GCE accelerators. At Leadership Roundtable, Peter set a specific miss target (3-6 items) for April's ~50 engineering deliverables. In Andrew's 1:1, Peter reinforced the Order of Operations prioritization model using CentOS Bridge as a concrete example and committed to improving team bandwidth visibility in Nathan's Monday meeting. Steve Wallace announced rotator cuff surgery (out Mon-Fri). TPS report v2.0.0 shipped with Justin's refactor into importable Python library.
Wins
- Strategic email to Manu sent same-day with Bjorn review — fast execution on relationship repair - GCE relationship strong: Hassan preparing NVIDIA VP pitch with explosive Rocky Linux growth data - Thermo Fisher agreed to AWS Fuzzball POC — new pipeline opportunity - TPS Report v2.0.0 shipped with 261 tests, importable Python library (Max + Justin collaboration) - Google GCE March ADCH numbers: every month in Q1 2026 beat 2025
Challenges
- Google GDC relationship at critical point — 90-day termination-for-convenience clause - ~50 April deliverables list too large, contains mis-categorized items and placeholder dates - CentOS Bridge customers at risk due to conscious deprioritization - Steve Wallace out for surgery this week — VP Engineering coverage needed - Multiple TPS items continuing to slip: RESF Board items 10d+ overdue, Fuzzball Observability 7d overdue - RLC+ 9 ISOs/Netboot confidence dropped to 0%