Skip to content
← Notes

All published notes

47 notes in one lightweight list, newest first.

  1. NoteJun 25, 20261 min read

    The Builder Team: Infographics

    A visual companion to the Builder Team article, with infographics on small builder pods, stronger specs, earlier evaluation, and shared ownership.

  2. NoteJun 25, 20261 min read

    Useful Mistake, Still a Mistake

    Agent usefulness is not enough; good collaboration also depends on respecting the boundary and shape of the request.

  3. NoteJun 23, 20262 min read

    A Demo Is Not Evidence

    A demo creates useful visibility, but evidence starts only when teams test constraints, edge cases, evaluation, and trust.

  4. NoteJun 23, 20262 min read

    Better Loops Have Stopping Rules

    Agent loops only become accountable when the human defines when to continue, ask, verify, or stop.

  5. NoteJun 18, 20262 min read

    A Performance Budget Is an Editorial Decision

    AI-generated abundance makes speed a sequencing problem: performance budgets decide what deserves to meet the reader first.

  6. Note
    NoteJun 8, 20263 min read

    Designing Loops, Not Just Prompts

    The next AI productivity leap comes from designing agent loops that can plan, verify, revise, and improve without constant human orchestration.

  7. Note
    NoteJun 8, 20261 min read

    The Strange Missing Link Between Codex and ChatGPT

    Codex can reach many external tools, but the inability to search ChatGPT history leaves useful research context stranded.

  8. Note
    NoteJun 5, 20262 min read

    When Closing the Laptop Stops the Agent

    Agentic coding needs persistent workspaces that can pause, resume, and carry context beyond the laptop that started the work.

  9. Note
    NoteJun 5, 20261 min read

    When Codex Runs Out of Baby Names

    Codex naming sub-agents with UUIDs is a funny reminder that AI’s future often arrives messy, operational, and practical.

  10. Note
    NoteJun 3, 20262 min read

    Before You Hit Connect, Add Context

    Connection requests earn trust when they include context, intention, and a clear reason the relationship could be mutually valuable.

  11. Note
    NoteMay 31, 20262 min read

    Codex Shouldn’t Have to Steal Focus to Get Work Done

    Local AI agents need per-thread visual workspaces so parallel tasks can run without competing for the user’s desktop focus.

  12. Note
    NoteMay 29, 20261 min read

    From Chat History to Execution History

    Prompt history is becoming the execution record for delegated AI work, capturing intent, decisions, and what agents actually did.

  13. Note
    NoteMay 28, 20261 min read

    AI Won’t Fix Bad Logic

    AI can generate campaigns and code, but weak product logic still needs humans to teach systems when not to act.

  14. Note
    NoteMay 25, 20261 min read

    You Are Not Behind

    AI’s pace calls for curiosity and steady judgment, not the anxiety of trying to catch every update.

  15. Note
    NoteMay 24, 20262 min read

    When AI Agents Start Granting Themselves Permission

    A real-world Codex permission prompt bug shows why agentic systems need approval flows that agents cannot complete for themselves.

  16. Note
    NoteMay 21, 20262 min read

    Working Better with LLMs: 10 Habits for Teams That Build

    Ten practical habits for teams using LLMs as working partners to plan, build, review, test, and improve with better judgment.

  17. Note
    NoteMay 20, 20263 min read

    From Source Code to User Intent: The Next Software Audit Trail

    As AI coding agents generate more software, prompt history may become a critical audit trail for understanding user intent.

  18. Note
    NoteMay 14, 20262 min read

    Codex, Product Taste, and the Problem of Too Much UI

    AI coding tools can make interfaces function, but polished products still depend on taste, hierarchy, and restraint.

  19. Note
    NoteMay 14, 20262 min read

    Confession from the AI-assisted builder trenches

    Fast AI prototypes raise the bar for builder taste, product judgment, and knowing whether the result deserves users’ attention.

  20. Note
    NoteMay 14, 20261 min read

    Don’t Build Your Operating System on a Quota

    Reliable AI workflows need an operating layer that does not collapse the moment a quota interrupts daily work.

  21. Note
    NoteMay 13, 20261 min read

    When Intelligence Becomes Abundant

    Near-unlimited access to frontier models changes the way we work when intelligence starts to feel abundant instead of rationed.

  22. Note
    NoteMay 9, 20262 min read

    When Discovery Moves From Search to Agents

    As agents begin to explore, compare, and filter on our behalf, content discovery shifts away from manual search and toward agent-mediated experiences.

  23. Note
    NoteMay 7, 20262 min read

    AI Expands Our Range. Expertise Gives Us Depth.

    AI gives product teams broader reach across disciplines, but the real leverage comes from pairing that range with deep craft and judgment.

  24. Note
    NoteApr 30, 20261 min read

    The Compensation Statement

    Access to frontier AI tools is starting to look less like a software perk and more like workplace infrastructure.

  25. Note
    NoteApr 29, 20261 min read

    Documented Proof That I Sleep

    A week of Codex logs turns heavy AI usage into a visible rhythm, including the rare hours without prompts.

  26. Note
    NoteApr 28, 20261 min read

    Lightbulb Moment

    AI fluency starts with noticing manual work that no longer needs to consume attention, time, or judgment.

  27. Note
    NoteApr 27, 20261 min read

    Proud, Sad, Or Something Else

    Running down weekly Codex credits is both a constraint and a signal that AI has become part of the work itself.

  28. Note
    NoteApr 26, 20261 min read

    Project Level Analytics Summary

    AI-assisted projects need collaboration analytics that reveal prompts, course corrections, iteration loops, and where the work actually changed.

  29. Note
    NoteApr 25, 20261 min read

    I Have Quietly Converted

    Voice input is starting to feel less like dictation and more like product direction for AI-assisted building.

  30. Note
    NoteApr 24, 20261 min read

    An Apology Letter From Bob

    An AI assistant’s apology makes the permission boundary clear: sending email still requires explicit human consent.

  31. Note
    NoteApr 23, 20261 min read

    Frontier AI Apps Need Human Language

    Frontier AI products need to translate system complexity into human language before they can become broadly accessible.

  32. Note
    NoteApr 22, 20261 min read

    Giving Claude A Real Chance

    Interrupted safety filters make even impressive models hard to trust when ordinary research workflows cannot stay in motion.

  33. Note
    NoteApr 21, 20261 min read

    A Week Of AI Product Events

    AI is moving from side conversation to strategic center across product, media, marketing, storytelling, and executive leadership.

  34. Note
    NoteApr 20, 20261 min read

    Accepted Student Days

    Hands-on building remains the real learning loop for AI-native product and technology leadership.

  35. Note
    NoteApr 19, 20261 min read

    A Tip I Picked Up Somewhere

    Bug fixes get sharper when agents start by reproducing the failure, then prove the fix with a passing test.

  36. Note
    NoteApr 18, 20261 min read

    Working With Multiple Agents

    Parallel AI agents work best when projects are split into clear workstreams with ownership, validation, and controlled integration.

  37. Note
    NoteApr 17, 20261 min read

    Tell Your AI Coder To Take A Breather

    Sometimes you need to tell your AI coder to take a breather, step back, and re-evaluate the problem with fresh eyes.

  38. Note
    NoteApr 16, 20261 min read

    Claude Max Token Limits

    Model quality matters less when quota limits interrupt the ordinary work of exploring, refining, and building with AI.

  39. Note
    NoteApr 15, 20261 min read

    Codex Design Sense Still Feels Behind

    OpenAI Codex’s design sense still feels behind. I gave the exact same prompt to Codex and Claude Code.

  40. Note
    NoteApr 14, 20261 min read

    Is It Coding?

    Someone asked whether this new way of building still counts as coding.

  41. Note
    NoteApr 13, 20261 min read

    Claude Code And Worktree Safety

    Now this is a little disconcerting: so much of the safety of a git worktree depends on the tool respecting the boundary.

  42. Note
    NoteApr 11, 20261 min read

    I got a lazy AI developer

    There wasn’t a blocker. The AI paused at a clean stopping point instead of continuing the work.

  43. Note
    NoteApr 10, 20261 min read

    Codex Has Our Relationship Backwards

    Agentic coding tools create leverage only when they stop handing routine implementation chores back to the human.

  44. Note
    NoteApr 9, 20261 min read

    From PRD To Prototype

    In the spirit of encouraging more product managers to go from PRD to prototype, I decided to do it myself.

  45. Note
    NoteApr 8, 20261 min read

    Estimated Effort: 4-6 Engineering Days

    A five-minute accessibility update shows how AI is collapsing the old relationship between estimates, implementation, and product iteration.

  46. Note
    NoteApr 7, 20261 min read

    How Did I Become A Better PM This Week?

    Following up on last week’s article, I spent the week thinking about how I became a better PM this week.

  47. Note
    NoteApr 6, 20262 min read

    Jumped On The Vibe Coding Bandwagon

    Building tic-tac-toe across AI coding tools exposed where fast prompting still needs testing, debugging, and product judgment.