1. Use the right model for every task

    (you can't) Running models is expensive and we have this dream of being able to "use the right model for every task", so we can be cost-effective. The problem is: How do you determine which is the right model for a specific task? What does it even ...

  2. I have not written a line of code in five months

    Some honest notes after several months building software with LLM agents: what works, what breaks, and why architecture, tests, context, and taste matter more than ever.

    • AI
    • SOFTWARE-ENGINEERING
    • AGENTS
  3. Single-threaded async Rust

    A small exploration of Rust futures, executors, and why async Rust can run perfectly well on a single thread when the runtime is built that way.

    • RUST
    • ASYNC
  4. URL Query String as SPA Initial State

    A short 2019 note about keeping SPA filter state in the URL query string so reloads and shared links show the same page state.

    • SPA
    • URLS