One question. One answer. Built to be cited.

How the core PM artifacts and rituals are being rewritten for the AI era.

• Is llms.txt enough to make a site useful to AI assistants?llms.txt is an index, not content. Anthropic, Vercel, and Supabase publish one. What the 780+ sites doing it are still missing, and what to ship next.
• Why is a polished mockup no longer evidence of conviction?Fidelity was a proxy for how much thinking sat behind the work. AI broke it, so the most finished-looking work now gets the shallowest critique.
• Should you be the platform underneath or the application on top?Glean and Alation both retreated to the substrate. You cannot own a daily habit and be neutral plumbing at once. Decide by renewal mechanism.
• What do designers do when AI generates the screens?Drawing got cheap, deciding did not. The designer's job moves from producing artifacts to specifying behavior: constraints, ranges, rubrics, and the wrong path.
• How do you level and promote PMs when the deliverables are gone?When AI collapses the documents PM promotions ran on, leveling has to measure judgment. What counts as evidence at each level, L4 to Principal.
• What does it cost to land a product versus build it?Building costs collapsed, landing costs did not. Where each cost lives, who pays, when it hits, and why orgs fund the cheap half and starve the other.
• What does it mean to land a product, not just ship it?Shipping is code in production, launching is the announcement, landing is durable behavior change with realized value. The definition and four failure modes.
• Who owns product landing in an organization?Landing falls in the seam between product, marketing, sales, and success, each handing off before adoption. Why it is nobody's job, and three fixes.
• Why does Whatnot run product with so few PMs?Whatnot runs an $8B GMV marketplace with ~22 senior PMs, hired 1 of 31,832 applicants, and maps PMs to problems, not pods. Never hire by default.
• Why is the PM career ladder breaking in the AI era?The PM ladder runs on artifacts as promotion evidence. When AI collapses the document, it removes the currency the ladder was calibrated on.
• If building is cheap now, what is the real constraint on product builders?Building stopped being the bottleneck. The 2026 CPO report shows speed to market became the top CPO challenge. The constraint moved to landing, not launching.
• Does AI delete the director of product?AI does not delete the PM or the VP first. It exposes product middle management, the Director and Group PM whose main job is moving information up and down.
• Does continuous discovery scale for AI-native products?Teresa Torres' continuous discovery fits human-centric SaaS and breaks for agent products. Punctuated discovery is the alternative: 60/40, three-week sprints.
• How do you hire a Builder PM?The old PM hiring loop tests presentation and frameworks that LLMs made easy to fake. Hire by what a candidate can ship in four hours, not how they talk.
• How does AI change the PM-to-engineer ratio?The 1-to-7 PM ratio was never about engineers. It priced how much build output one PM could direct. AI raised each engineer's output and broke the math.
• How is the product manager role changing with AI?AI-forward companies hire Product Builders, not spec-writers. The PM role now rewards six skills led by rapid prototyping and customer proximity.
• Is Agile dead, and what replaces it?Agile was a coping mechanism for slow, costly building. AI killed both founding constraints. What replaces it is not waterfall, it is the spec-first loop.
• Is being a Product Builder about coding?No. Product Builder was never about a PM learning React. It is about collapsing the distance between deciding and making, and killing the coordination tax.
• Is being AI-first a product decision or a tooling one?Being AI-first is a product decision, not a tooling one. Using AI runs your old process faster. Being AI-first rebuilds the process as if AI was there day one.
• Is product management dead?Product management is not dead, it escaped the product team. Every function deploying AI workflows is structurally a product team that needs builders.
• Is Scrum obsolete for AI-native teams?Scrum is obsolete and the market is voting with hiring. The replacement has a name: a four-to-six person builder pod with no ceremony tax.
• Is the PM-as-translator role dead?For 15 years PMs were professional middleware, reformatting customer input into decks for engineers. AI and MCP stripped away three layers of translation.
• Is the PRD dead?The PRD is dead because building got cheap. A clickable prototype, a one-pager, a 5-minute Loom, and a 30-minute conversation replace it.
• What is the difference between a PM and a Product Builder?A PM shipped documents and ran handoffs. A Product Builder ships prototypes, writes evals, and owns a surface end-to-end. The line-by-line difference.
• What belongs in a PRD versus a spec versus a brief?A PRD did three jobs badly. Split them: the brief carries the why and the bet, the eval carries the commitments, the prototype carries the solution.
• Should you call yourself an AI PM?No. AI-native describes tool familiarity, not decision ownership, and it will not survive resume screening. Name the outcome you own, like I own activation.
• Should you merge the PM and product owner roles?Merging PM and product owner fixes the symptom, not the cause. The real fix is killing the Scrum frame the PO role exists to staff.
• What are the three artifacts that replace a PRD?A one-page brief, a throwaway prototype, and an eval. Each does one job the PRD did badly: memory, alignment, and commitment. With the templates.
• What do boards expect from product in 2026?The 2026 CPO board deck is seven slides: outcome ledger, unit economics, eval scorecard, agent inventory, cycle time, headcount-to-output, downside.
• What is a one-page brief and what goes in it?A one-page brief replaces the PRD with six sections: problem, evidence, solution with a prototype link, metrics, risks, and the ask.
• What is an AI product engineer?An AI product engineer combines PM strategy, design, and engineering, with AI amplifying each. One person now does what used to take a team of three.
• What is dual transformation for a product org?Dual transformation runs a legacy SaaS and its agent-native successor on two clocks: two cadences, three talent categories, six CEO numbers, three rituals.
• What is service-as-software?In service-as-software, AI agents do the work and humans review and handle edge cases. It changes UX, reliability, pricing, and the PM role itself.
• What is substrate-first engineering?Substrate-first engineering means engineers build the platform that lets non-engineers ship safely, not the features. Four pieces make it up.
• What is the AI noise tax?The AI noise tax is the credibility cost of three patterns: superpower LinkedIn posts, bolted-on AI sidebars, and PMs arguing prototyping breaks empathy.
• What is the AI product operating model?The AI product operating model keeps empowered teams and continuous discovery, but agents kill the PRD and compress discovery from four weeks to four days.
• What does the org chart for an AI product company look like?AI coding tools raised developer output but org velocity stayed flat. The new org chart fixes three layers: adoption, review, and the coordination middle.
• What does a Product Builder job ladder look like?A four-level Builder PM ladder from L4 to Principal. Five dimensions scale with level; the operating model does not: prototypes, evals, user validation.
• What replaces the product org by 2028?Ten checkable predictions for product orgs by mid-2028: median team of six, code in week one, evals the top skill, hybrid pricing, extinct APM programs.
• What should a PM job description look like in 2026?I audited 200 PM job descriptions. 196 could have been written in 2022. The four tells of a dead JD, the four tells of a live one, and how to rewrite.
• Which software categories is AI killing?Three forces are collapsing software categories: search becoming answers, tools becoming agents, subscriptions becoming outcomes. A PM's field guide by bucket.
• Why does distribution beat product in the AI era?AI made product cheap to build, so product quality gets copied in an afternoon. Distribution, the channel and brand and install base, is the only moat left.
• Will AI replace product managers?No, but AI will replace 80% of what most PMs do. The surviving 20% is customer empathy, judgment, leadership, and vision, and it becomes everything.

What to stop doing, and how to run the kill so the org actually lets go.

• Which design rituals should you kill, and what replaces them?Nine design rituals that produce a receipt instead of a decision, each with its replacement, plus the four-step protocol that makes a kill stick.
• Are OKRs still worth it for AI products?OKRs were built for weekly builds and monthly signal. For AI products with daily signal and four-hour prototypes, replace them with a Rolling Outcome Ledger.
• How do CPOs win the AI transition without stalling?The soft pivot is a fantasy. Win with a sunset date, rewritten comp, a pre-sold gross-margin trough, and the seven-decision sequence run in order.
• How do you decide whether to cannibalize your own product?Four diagnostic questions route the call to sunset, refresh, or split. Architecture, ICP, gross-margin trough, and buyer decide, not comfort level.
• How do you deprecate a feature?Deprecate on signal, not politics. Write a kill condition at launch, pick one of three tiers, and send individual notices to users from the last 90 days.
• How do you kill a feature without breaking trust?Kill a feature by running discovery on non-adopters first, then reorganizing before you rip out code. One kill lifted platform activation 15%.
• How do you run empowered teams when the CEO doesn't buy in?You don't need permission to run empowered teams. You need evidence. Four tactics, repeated for six months, until the results make the case for you.
• How do you sunset a pricing tier?Sunset a pricing tier over 18-24 months with a three-wave customer migration, a comp plan rewrite, and a post-sunset reorg. Five patterns kill most sunsets.
• Is spec-driven development (IDSD or SDD) worth it?SDD and IDSD are the same methodology layer with different stickers. Skip both. The prototype is the spec, the eval is the acceptance test.
• Should you kill AI office hours?AI office hours are 2026's agile transformation: high attendance, zero shipped output. Kill them and run paired shipping, eval reviews, and kill list reviews.
• Should you kill the PRD entirely?The PRD was a 2015 fix for a 2015 problem. Replace it with three artifacts: a working prototype, a one-page eval rubric, and a one-page README.
• Should you kill the roadmap?Kill the quarterly roadmap and run a one-page live bet portfolio instead: 5-7 active bets, each with a hypothesis, signal, kill condition, and decision date.
• Should you kill the status meeting?The status meeting exists because nobody trusts the dashboard. Replace it with one live product page and reclaim about six hours a week per PM.
• Should you still run an APM program?LinkedIn's CPO killed the APM program. Replace the two-year rotation with a 12-week Product Builder apprenticeship: four prototypes, one production ship.
• Were empowered product teams just a ZIRP feature?Empowered teams were a luxury of cheap capital, not a law. Expensive capital killed patience, AI killed the constraint. Autonomy survives, ceremony does not.
• What is a Kill List and how do you run one?A Kill List is the standing counterpart to a roadmap: a maintained list of things to stop. Run it on a cadence with one entry test: would we start this today?
• What is the anti-backlog?The anti-backlog replaces a bloated backlog with a live queue capped at two weeks, a hypothesis library, and a kill list. Migrate in four weeks.
• What should you do instead of a feature request queue?Replace the feature request queue with signal synthesis, prototype-first investigation, and a seven-item outcome bet ledger. Your 847-item backlog is debt.
• What is wrong with the 'I built it in a weekend' flex?Speed to demo is a vanity metric. The weekend-build flex measures the cheapest, most brittle thing you make. Ask if it was still running the weekend after.
• Why can killing a product beat launching one?A good kill often creates more value than a mediocre launch, but the org is blind to it. Killing well is the rarest and most valuable product skill.
• Why is a reorg a product decision?Conway's Law means your org ships its communication structure, so a reorg sets your roadmap ceiling before a ticket exists. The org chart is a product spec.

Reusable agent blueprints and the plumbing that makes them reliable.

• What should I put in a Claude Project for product management?Definitions add nothing a model already has. Put in positions, artifacts, and links. What actually changed the output, and the file to start with.
• What is the difference between an AI agent, a workflow, and an automation?Automations follow rules, workflows add a model at fixed steps, agents decide what to do next. Here is how to tell which one you are actually building.
• Do different types of PM need different agent stacks?Yes. Technical, Growth, and Platform PMs need different agent stacks. The Product Health agent is shared, but configuration matters more than capability.
• How many AI agents does a PM actually need?Seven. One agent per stage of the PM operating system, each with a named owner, a single signal stack, one output surface, and a quarterly kill-switch test.
• How do you put stakeholder updates on autopilot?An agent pulls data from six systems and drafts three audience-tailored updates every Monday. Saves 5 to 6 hours a week. Keep manual review for 3-4 weeks.
• How do you build a competitive intelligence agent?Wire an agent to seven data sources, run it every other Monday, and get one 10-minute intel report. The customer signal is what manual research misses.
• How do you build a daily focus agent for PM work?An AI chief of staff reads your calendar, Slack, Jira, and email at 7 AM and delivers the three things that actually matter today, ranked by a scoring model.
• How do you catch metric drops automatically with an agent?An agent checks your KPIs hourly, detects drops outside the noise band, correlates likely causes, and ships a clickable prototype fix before you read the alert.
• How do you generate launch comms across channels with an agent?An agent reads a shipped feature's full context and generates six channels of GTM copy in one pass, in under 4 minutes, all tagged pending review for PMM.
• How do you build a PRD generator agent?Wire an agent to your opportunity stack, research repo, and architecture docs. It drafts a research-grounded PRD in eight sections, weekly on a schedule.
• How do you build a product health dashboard agent?A weekly agent pulls 3 weeks of analytics and analyzes adoption curves, cohort retention, engagement depth, and performance to shape quarterly strategy.
• How do you catch product red flags early with an agent?A morning agent scans support, incidents, backlog, Slack, and CRM, then posts one triage report in three tiers. Reclaims 40+ hours a quarter.
• How do you build a release readiness agent?A weekly agent audits PRDs, release notes, ETAs, feature flags, and GTM readiness, then gives a go/no-go. By week three, launches are boring.
• How do you run a retrospective with an agent?A Friday agent reads your retro notes, compares them to the last 8 retros, finds recurring themes, and tracks action-item follow-through.
• How do you build a roadmap progress agent?A daily agent cross-references your roadmap against actual commits, so you see the drift weeks early instead of a week before launch.
• How do you automate sprint planning with an agent?A Monday agent turns prioritized opportunities into a Jira-ready sprint plan with story points, capacity fit, dependencies, and a confidence score.
• How do you turn support tickets into product signal with an agent?A daily agent clusters your support queue by root cause and segment, compares against baseline, and flags spikes over 25% before they become crises.
• How do you prioritize tech debt with an agent?A weekly agent analyzes four sprints of Jira data to quantify which debt slows velocity, then ranks fixes by impact times urgency over effort.
• How do you build a win/loss analysis agent?A bi-weekly agent reads every won and lost deal and extracts seven categories of product insight, so you build features that move deals.
• How do you turn a support ticket into a reviewable PR automatically?An agent watches Zendesk and Sentry, reproduces the bug, locates the broken code, writes the fix plus a regression test, and opens a reviewable PR.
• How do you turn a customer request into a prototype in minutes?Wire an agent to your signal channels, design system, and codebase. It generates a clickable prototype from a customer request in about 7 to 10 minutes.
• How do you build a customer interview synthesis agent?An agent reads every interview transcript from the week and ships one theme report in minutes. Signal only emerges across 8+ interviews, not one.
• How do you build an OKR tracking agent?A daily agent scores every key result against the trajectory needed and predicts weekly confidence, so you catch slippage at week 5, not week 13.
• How do you prioritize opportunities with an agent?A weekly agent reads all five DISCOVER agent outputs and produces one prioritized, OST-ready opportunity stack with confidence, impact, and dependencies.
• How do you connect Claude to your data sources for PM work?Wire Claude into your real work tools via MCP servers, schedule agents on cron, and pipe results to Slack. The full setup takes about two hours.
• How do you measure agent reliability without gaming it?Grade every agent on four metrics: precision, escalation rate, time-to-output, and trust trend. Escalation rate is the one nobody tracks.
• How do you measure PM velocity from signal to ship?An agent tracks every product change across the seven PM OS stages, computes time-in-stage, names the weekly bottleneck, and flags stuck items.
• Should PMs vibe code, and what should they build?Yes, PMs should vibe code with Claude Code, but for internal tools and prototypes, not customer features. Same skill, completely different blast radius.
• What Claude skills should every PM build?The five Claude Skills to build first: Interview Synthesis, PRD One-Pager, Stakeholder Update, Competitive Intel, and OKR Tracking. 30 minutes each.
• What happens when you hold AI agents to a performance review?Run a quarterly performance review on your agent fleet using four metrics. Last cycle, three of my agents failed and I shut them down. Here is the scorecard.
• What is agent-to-agent dispatch?Agent-to-agent dispatch is AI agents handing work to other agents with no human relaying context. A listening agent detects, a build agent prototypes.
• What is an agent operator team?An agent operator team supervises the AI agents that build the product. At falkster.ai six of them are pets, paid in treats, and never miss an anomaly.
• What does it mean to manage a fleet of agents as PM-as-editor?PM-as-editor is the skill you need once your agent fleet runs: read output like a senior PM reads a PRD, use a four-tier trust ladder, feed edits back.
• What is prompt ops?Prompt ops means treating prompts like production code: version control, PR review, evals on every change, staged rollout, and one-click rollback.
• What is the PM agent stack?A personal PM agent stack layers open-source Claude tools on Claude Code, mapping 18 tool categories to the 7-stage PM operating system.
• What is the seven-agent PM fleet?A lean PM AI fleet of one agent per operating-model stage, each with a named owner, one signal stack, one surface, and a quarterly kill-switch test.
• How do three agents collapse an eight-week cycle to ten days?Three AI agents (Instant Prototype, Auto Bugfix, Launch Comms) turn a customer Slack DM into a shipped, announced feature in ten days instead of eight weeks.
• What replaces the product triad and pods in an AI org?A human department doing judgment work coordinating with an agent department doing routine and synthesis work, on exception via a published interface.
• When should you not use AI in a product?Before wrapping a surface in an LLM, walk a five-step decision tree: can a rule, a query, a form, or a heuristic do it first. The hybrid pattern halves cost.
• Which AI prompts replace the most PM busywork?The three highest-leverage AI prompts for PMs: the JTBD extractor, the support-ticket theme clusterer, and the one-pager-from-notes generator.
• Why do AI agents fail in production?Three patterns dominate: distribution shift the evals missed, correctness that was beside the point, and cost-compounding autonomy with no human gate.

Decisions from real launches, with the numbers and the tradeoffs named.

• Why did SpaceX buy Cursor for $60 billion?SpaceX closed its $60B all-stock buy of Cursor on Aug 14, 2026. It bought the fastest-growing product in AI coding, and a supply-chain fix for its one weakness.
• Why is Harvey AI worth $11 billion?Harvey hit $350M ARR and an $11B valuation selling AI to Big Law. The valuation is a bet on vertical depth, the workflows and trust a model can't copy.
• Why is Microsoft's agent called Scout instead of Copilot?Microsoft named its agent Scout, not Copilot, because it gets its own Entra identity and acts as a separate principal. The name follows the architecture.
• How can a prototype beat a spec in a product decision?A prototype settles the argument a spec can only prolong: it produces evidence from real users instead of compounding the team's guesses. Here is the mechanism.
• What can an eval catch that a demo cannot?A demo shows a chosen best case. An eval shows the distribution, including the failing slice a demo cannot reveal. The one that almost shipped.
• Which companies treat product landing as a real discipline?Snowflake solved landing ownership with pricing, not org design: it only gets paid when customers use the product. How consumption pricing forces it.
• Whose job does the Product Builder shift threaten, and how do you lead them through it?The shift threatens the person whose craft AI made cheap, often your best writer. How to have the honest conversation early, and why delay is the real cruelty.
• Why do successful launches fail to land?A launch can hit press, signups, and week-one activation and still flatline by month two. The three places landing falls through, and how to instrument.
• Can you run a staff meeting off a live dashboard?I ran my product staff meeting off an agent-refreshed dashboard for a quarter. What it killed, what it surfaced, and the one mistake I would not repeat.
• Can a prototype kill a project before you build it?Yes. A two-hour prototype and five customer calls showed a feature was structurally wrong 48 hours before kickoff, saving 600 engineering hours.
• How does a PM build financial fluency?Financial fluency is now a PM requirement: the five numbers to know cold, a 90-minute session to rebuild cost per outcome, and a four-week plan.
• How do you lead a product team through layoffs?The cut is not the hard part. Keeping the survivors building is. What survives a layoff, and the three leadership moves that keep the team from quietly rotting.
• How do you move from SaaS to service-as-software as a CPO?Moving from SaaS to service-as-software means rebuilding pricing, product, and the org chart around outcomes. A CPO field guide with the decision framework.
• How do you run a CPO listening tour that works?Replace the 40-coffee-chat listening tour with a listening ledger: nine fixed questions, falsifiable claims, contradiction tracking, and a Friday synthesis.
• How do you run a QBR that forces decisions, not applause?A 9-slide QBR structure that makes the quarter legible and the next quarter decidable in 30 minutes: scoreboard, decision review, kills, bets, and anti-bets.
• How do you write an exec update that lands?An exec update read in 90 seconds: a five-number scoreboard, three decisions with evidence, one ask with a default, and one line on what you killed.
• How do you defend a margin drop to your CFO?The three-session CFO conversation that turns a 20-point gross margin trough into a managed plan: the trough math, the comp set, seven indicators.
• How should you price an AI product?Per-seat is dead for AI. Price the work the seat is no longer doing, with hybrid, outcome-based, tiered, or usage models on a unit customers understand.
• How do you run a pricing migration?The six-quarter playbook for per-seat to outcome pricing: four mandatory conversations, three migration waves, the margin trough, and the recovery.
• Is gross margin a product manager's job now?Yes. Cost per successful action is the new primary PM metric. Three levers earn back the margin: model routing, prompt hygiene, and caching plus early exit.
• Is taste the last moat in product?When execution costs almost nothing, the scarce input is judgment about what to build. Taste is a balance-sheet asset and the one thing a CEO cannot delegate.
• Is the product budget now a compute budget?In an agent-native org, output scales with compute and judgment, not headcount. The product budget starts behaving like infrastructure. What CPOs must change.
• How do you tell a reversible product bet from an irreversible one?Ask one question: group stage or knockout? Reversible bets survive a bad week, irreversible bets churn the account with no replay. Downside exposure.
• What breaks in a 90-day SaaS-to-agents transition?Three things break in the first 90 days of a SaaS-to-agents transition: the CFO conversation, the lead customer legal review, and maintenance team morale.
• What breaks when you move off per-seat pricing?A field report from the day a per-seat tier sunset: migration drift, a churning cohort, a loose unit definition, and disputes at 12x projection.
• What does a $6.5B startup exit actually teach you?A big exit teaches you less than a failure does, because a win hides which of your decisions were good. The real product-leadership lessons live in the losses.
• What do acquirers actually buy in a startup acquisition?Acquirers rarely buy your product or revenue. I sold MVC to Microsoft. They buy one of four things: a capability, a team, a defensive block, or time.
• What do founders actually want from a CPO?Founders do not want a CPO who replaces their product judgment. They want it cloned and scaled without losing control. Why it fails, and how to win.
• What does an AI-native CPO do in the first 90 days?The AI-native CPO first-90-days plan: audit reality not the deck, instrument cost and quality, then kill dead ceremonies and defend the calls only you own.
• What goes in the product section of a board deck?Seven slides in ten minutes: scoreboard, what changed, bets table, kill list, the one risk, and asks. No feature parade. What a board can actually act on.
• What goes in a CPO 30/60/90 plan?A CPO 30/60/90 built as an audit, not a tour: four audits, a trust ledger, a coalition map, and a day-90 readout the board will quote back for a year.
• What happens to your product after an acquisition?After acquisition your product often gets killed, and that can be correct. Microsoft killed the product I built after buying MVC. Why it was right.
• What is per-outcome pricing?Per-outcome pricing ties the price to a unit of work delivered, not access to a tool. Examples, margin math, and the five contract terms it needs.
• What is the Jevons cliff in outcome pricing?The Jevons cliff: inference costs drop ~50% a year, so your outcome price drifts from its sweet spot until a competitor undercuts you. Hold, pass, or split.
• What five questions make a product review worth the time?Most executive product reviews are status theater. Five questions convert them into decision-forcing sessions that surface judgment, cost, and risk.
• Why do acquired products die inside big companies?Acquisitions fail in three steps: the team loses its constraint, inherits parent process, and stops shipping. I watched it three times inside Salesforce.

Methods from practice: evals, discovery, measurement, prioritization.

• What is Teresa Torres's three-step approach to AI evals?Error analysis, choose the eval type, run the loop. What the Product Talk guide says, what to steal first, and the step that turns an eval into a gate.
• Do forward deployed engineers scale, or is it just consulting?Forward deployment costs like services and has to earn like software. When the math works, the buyer's lock-in objection, and the signal to stop sending engineers.
• Does zero data retention mean the vendor keeps nothing from your AI agents?Salesforce promised at Dreamforce 26 that your data does not go in the models. True for the model layer. Here is what that promise does not cover.
• How is a forward deployed engineer different from a sales engineer?One line separates them: a sales engineer's job ends when the customer signs and an FDE's job starts there. Confusing the two is behind most stalled AI pilots.
• What is a forward deployed engineer?The role Palantir invented and OpenAI, Anthropic, Sierra, and Google Cloud hire for in 2026. What it is, what it is not, and why it is a Product Builder in disguise.
• When can you actually enforce a rule on an AI agent?Spotify could gate file reads over 350 lines and not boilerplate writes. The three-question test that separates rules you can enforce from documentation.
• What is Marty Cagan's fresh definition of the product role?Cagan maps Benedict Evans's three product skills onto problem discovery, value risk, and viability risk. What it says, and what the mapping leaves out.
• Can you retrofit context for AI agents?No. Context is the only agent layer captured at write time. Control planes, routing, and governance can be added later. The reasoning behind a decision can't.
• Why did Miro sell for only 2.3x ARR?Miro is profitable with ~$600M ARR and sold at 2.3x. Compression explains the range, not the position. The mechanism is legibility.
• What did Salesforce's Q2 FY27 earnings say about agentic AI?Adjusted EPS matched consensus. GAAP jumped on an Anthropic gain. And Salesforce shipped its CRM inside Claude. The three numbers that matter.
• What are bottom-up evals, and why can't an LLM write them?Top-down evals restate the brief, so a model drafts them well. Bottom-up evals come from reading real outputs. Here is the conversion rule.
• Why do customers buy your product but not renew it?The reason a buyer approves you and the reason users keep coming back are different things. Most teams instrument only the first one.
• How do you design a screen that is generated at runtime?Specify the range, not the screen: best case, worst acceptable, and what the product refuses to do. Everything between the last two ships without review.
• How do you write a design rubric that scores work without you?Extract the rubric from thirty graded outputs, keep only dimensions with a named cost, score binary with severity, then calibrate on five outputs.
• What do you do in the thirty seconds after your AI is confidently wrong?Six steps in order: detect, admit in place, contain, correct visibly, name the prevention, adjust the confidence posture. Written before you need it.
• Does the IKEA vs Salesforce story prove AI automation fails?No. The viral story is quoted as proof AI fails, but both halves are stories of AI working, and the comparison misreads the tech, the 47%, and the sourcing.
• What are malleable loops in software?Dave Killeen's term for improvement moving through the software itself: the product self-checks, files scrubbed defect reports agent-to-agent, and ships.
• What is outcome-based pricing for AI agents and how does it work?Pay per resolved case, escalations free, tokens irrelevant. How Sierra's outcome pricing works mechanically, why it beats usage pricing, and where it breaks.
• Why are SaaS revenue multiples collapsing in the AI era?Airtable, Chegg, Stack Overflow, and Smartsheet repriced for two reasons. How to tell an AI repricing from a rate repricing, and which one you face.
• Why is Sierra AI growing so fast?Sierra hit $100M ARR in seven quarters and ~$200M by mid-2026. The mechanism is six operating choices, led by outcome pricing that ties revenue to landing.
• Why did Airtable sell for only 2.7x ARR?Airtable's ~$480M ARR grows over 20%, yet it sold at 2.7x revenue. AI didn't take the revenue, it took the option. Here is the mechanism and the fix.
• Are prompt logs the new switch interview?For what customers want and where your product fails, prompt logs beat the call: their own words, the exact moment, everyone. Interviews win on why.
• Can you trust your product intuition?Less than you think. Intuition is pattern-matching on a world that may no longer exist, and AI rotates the patterns faster than any gut can recalibrate.
• Does jobs-to-be-done still work in the AI era?JTBD assumes the underlying job is stable, and AI absorbs jobs faster than roadmaps react. It still names the job, it stopped saying how long it holds.
• What should you measure weekly versus quarterly for an AI product?AI features iterate ten times a week but outcomes take weeks to attribute. Measure direction weekly with leading indicators, outcomes on a slow cadence.
• How does a PM ship their first pull request with AI tools?The four-level PM PR ladder, PLANNING.md files in git, and the two non-negotiables that earn engineering trust before your first PR lands.
• How do you build a PM second brain from meeting recordings?PMs forget 90% of meeting context within a week. A four-step pipeline turns every recording into a searchable knowledge base that compounds over time.
• How do you do discovery when your customer is an AI agent?Agents don't answer interviews. Six methods replace the playbook: agent telemetry, failure-mode interviews with operators, capability audits, and more.
• How do you run a 15-minute sprint retro that improves things?Flip the retro ratio. AI generates a Sprint Health Report before the meeting from git, Jira, and Slack, so 15 minutes goes to decisions, not data gathering.
• How do you run your first product trio?A product trio is PM, designer, and tech lead deciding as a unit. Two hours weekly: 50 minutes discovery together, then decide with committed ownership.
• How do you run a weekly review that keeps you shipping?A three-phase weekly review in 30 minutes instead of 3 hours: gather the brief with AI, triage with five decision questions, communicate a standup opener.
• How do you train product judgment deliberately?Judgment trains like a sport: reps plus a scorecard. Use a decision log, written confidence scored quarterly, premortems, and one-way versus two-way doors.
• How do you write for machines, not just executives?Half your readers are agents now. Swap prose PRDs for eval-set-plus-brief, version prompts like code, and score every artifact with the cold-read test.
• How do you write OKRs that don't suck?Most OKRs are disguised task lists. Use the behavior change test: if the key result were hit, how would customer, user, or business behavior actually change?
• How does a decision log train product judgment?A decision log separates decision quality from outcome quality, records confidence as a number, and calibrates judgment against what happened.
• How do you avoid survivorship bias in product management?You only hear from the users who stayed. Talk to churned users, track non-events, run feature post-mortems, and put churn drivers in prioritization.
• How do you build a working prototype in 60 minutes?The exact five-phase workflow: frame the problem, describe to Claude Code, iterate UX, add real data, deploy and share. A testable app in one hour.
• How do you build a signal map in your first 30 days as a PM?Skip the coffee chats. In your first 30 days, map the seven places truth enters the building and audit whether each gets captured, synthesized, and routed.
• How do you measure the cost of being wrong for an AI feature?Speed is free now. Triage decisions by reversibility, not size. Ship cheap-to-undo calls instantly and slow down on the ones you cannot reverse.
• How do you run a customer interview that actually works?Show up with full signal context, run the 30-minute prototype interview format, and let AI synthesize the transcript in minutes. Depth over volume.
• How do you run continuous discovery?Agents ingest every call, ticket, and NPS response, extract signals, and surface a ranked opportunity brief every Monday. Then you prototype same day.
• How do you test a product assumption in a week?Map assumptions Monday, build a prototype that tests the riskiest ones, test with five customers, decide Friday. The prototype is the experiment.
• What is downside exposure and how do you score a feature on it?Downside exposure asks how fast a silent 20% quality drop would cost you customers. Score on three axes: revenue, cost of a wrong answer, reversibility.
• What are direction metrics?Leading indicators measured on the cadence of the work itself. Seven signals that predict AI outcomes 4-8 weeks ahead, run on a two-layer system.
• What can no product framework teach you?Frameworks are training wheels. They teach the moves, but not judgment under ambiguity, which only comes from making real decisions and being wrong at scale.
• What does it mean to ship with observability?No feature leaves staging without the traces, metrics, and evals that tell you whether it works, before your first customer hits it. A seven-item contract.
• What does 'the eval is the spec' mean?The eval set replaces the PRD for AI features. 30 to 200 real input/output pairs define what good looks like, scored daily. The eval is the contract.
• What is an opportunity solution tree?An OST connects a business outcome to customer problems, candidate solutions, and experiments. Teresa Torres created it. AI populates it in minutes now.
• What is mob prototyping?One day a week, your PM, designer, and engineer build a working prototype together in one room. 24 person-hours instead of 41. Better output than solo.
• What is the impact loop?A four-beat operating rhythm that replaces sprints: Sense, Build, Measure, Amplify. It optimizes for responsiveness, not predictability. Same-day to eight days.
• When does vibe coding get more expensive than agentic engineering?Vibe coding starts cheap, then costs 3 to 10x more per feature past the crossover point. The CapEx versus OpEx cost curve, from Google's SDLC whitepaper.
• Which mental models break when building gets cheap?Four decision frameworks quietly price building as scarce and invert when AI makes it cheap: Opportunity Cost, Bottleneck, Sunk Cost, and Local vs Global.
• Why can't users describe what they need?Users know their problems deeply but prescribe solutions limited to what they know exists. Feature requests are symptoms, not specs. Show, don't tell.
• Why should you design the tournament instead of picking winners?Cheap prototyping did not kill product judgment. It moved the decision from picking the winner to seeding the bracket: which eight of fifty get built.
• Why do cheap prototypes sometimes kill good ideas?When building is nearly free, the demo replaces the argument. A rough prototype of a strong idea reads as a weak idea, and it dies in review.
• How do you write an eval rubric for an AI feature?Grade thirty real outputs on gut feel first, extract the rubric from your own disagreements, tie every dimension to a cost, and score binary, not 1-5.
• Should you build your own data governance for AI agents?You will not out-govern Alation, Collibra, or Purview. Plug into the governance stack the enterprise already has, and design three things now.