Writing

First-principles essays and shorter notes. Most pieces are also nodes in the field on the home page.

• noteMay 24, 2026

  Team memory needs permissions

  A shared agent system needs to know who can read, write, approve, and act before it becomes useful.

• noteMay 24, 2026

  Skills are bounded capabilities

  A skill should be small enough that its inputs, allowed actions, and failure modes can be named.

• noteMay 24, 2026

  Retrieval is a judgment layer

  Good retrieval is not just finding related notes. It is deciding what context is safe to use for the task.

• noteMay 24, 2026

  Raw intake is not memory

  The first job of agent memory is to capture without pretending the capture is already true.

• noteMay 24, 2026

  Memory promotion is the product

  An agent memory system becomes useful when it knows which records are allowed to become context.

• noteMay 24, 2026

  Feedback loops need a place to land

  A system only improves when corrections become memory, skill changes, eval cases, or routing rules.

• noteMay 9, 2026

  Why I love synthetic data

  It's the rare medium where breadth, creativity, and technical chops compound instead of trading off.

• noteMay 7, 2026

  Taste is the rare skill now

  Producing is cheap. Choosing is hard. The résumé that didn't get rewarded before is the one that does now.

• essayApril 28, 2026

  The immersive room is a listening environment

  We keep measuring whether immersive demos teach more. They don't. That's the wrong question.

• essayApril 21, 2026

  The apprenticeship was a curriculum

  Junior labor was the curriculum. AI eliminated the labor and didn't replace the curriculum.

• noteApril 16, 2026

  Synthetic data quality is not a number

  There's no general score, the absence is structural, and the test that matters is the one you run on your own use case.

• essayApril 14, 2026

  The author is still the author

  Use AI to draft, refine, brainstorm, code — fine. The moment your name is on it, every word is yours. 'GPT wrote it' does not transfer accountability. It just makes you a person who admits they didn't read what they sent.

• essayApril 7, 2026

  Dataset vs data product

  A dataset is a file. A data product is a contract about what questions it can answer, who owns the answers, and what happens when those answers change.

• essayMarch 31, 2026

  Is the AI-to-AI email loop the world we want?

  When my assistant writes to your assistant, what is the email actually for? Asymmetric drafting is one thing. Both sides automated is another, and the medium starts to lie about whose attention is being spent.

• noteMarch 26, 2026

  Real AI strategies are about subtraction

  If you can't name what you'll stop doing once the AI works, you don't have a strategy. You have a wishlist.

• essayMarch 24, 2026

  What 20 correlations taught me about consumer health data

  Twenty typed correlations, a sample-size gate per phase, and one confounder that broke half of them. The half that survived are the only ones I trust to surface to a user.

• essayMarch 17, 2026

  I thought AI could write the app. My team humbled me.

  Code generation has gotten very good. Naming the right abstraction has not. Most of what makes a system maintainable still happens before any code is written, and that part is not getting easier.

• essayMarch 10, 2026

  Cycle as a confounder

  Half the species moves through a four-phase hormonal cycle. Most consumer health correlations don't condition on which phase the data was collected in. Most of those correlations are wrong, or right for the wrong reason.

• essayMarch 3, 2026

  Zero-to-low sample generation

  A handful of real records, a regulator who won't release more, and a system that has to be tested anyway. Most synthetic data writing is for a different problem.