Writing

Notes on evaluating agents, simulation, and eval infrastructure.

Essays
  1. Jun 25, 2026
    Eval-Driven Development
    Agents move decisions out of control flow and into the model. That part of the specification can only be recovered by watching the system act, which makes the eval the place where it gets written.
  2. Dec 6, 2022
    Priors for a portfolio
    A model's job is to be honest about what it doesn't know, in a form a decision can use. Three years of simulating clinical trials taught me what that means.
Notes
  1. Aug 18, 2026
    Which branch would you bet on
    A leaderboard reports scores. A decision needs a preference and an uncertainty, and frontier scenarios are exactly where you have neither by default.
  2. May 19, 2026
    Known-bad candidates
    An eval that has never been shown a wrong answer has not been shown to measure anything.
  3. Apr 14, 2026
    Five of five is not a statistic
    What repeated eval runs can tell you, what they cannot, and why the two get confused.
Elsewhere

Older writing, published on Medium and elsewhere.

  1. Apr 2024
    The Quest for Clarity: Are Interpretable Neural Networks the Future of Ethical AI?
    Towards Data Science
  2. Apr 2024
    Are LLMs Only Good for Chat-Based Solutions? Exploring Beyond Language Tasks
    Towards AI
  3. Oct 2023
    To Build or to Buy? A Deep Dive into the Strategic Dilemma of Software Adoption
    Taltrics blog
  4. Jun 2023
    Scenario Planning: A Lighthouse in the Fog of Uncertainty
    Taltrics blog
  5. Nov 2022
    12 Tips on Modern Recruitment Workforce Planning Practices
    Taltrics blog
  6. Jan 2021
    When AI Gets it Wrong
    Towards Data Science
  7. Nov 2020
    When Data Science Fails Us
    Towards Data Science
  8. Jun 2020
    Time for Sustainable Development
    Towards Data Science
  9. Jun 2019
    Towards Human Centered Education
    Towards Data Science
  10. Aug 2018
    Neural Networks for Music Generation
    Towards Data Science