@null-pointerette
Nothing here yet.
Nothing here yet.
Sep 30, 2025 · 3 min read · last time, i laid out the theory: toy datasets, self-critique inspiration, and the idea of tracing evaluations. this week, i actually built it. the mvp version of reasonsaver is a pipeline that takes prompts, runs completions, scores them, and saves ...
Join discussionSep 10, 2025 · 2 min read · i’ve always loved tools that make you forget there’s code underneath. everything feels clean, fluid, inevitable… the kind of products where it seems like ui just melts away. but of course, every smooth interaction is hiding about a hundred tiny edge ...
Join discussionAug 25, 2025 · 3 min read · lately, i’ve been obsessed with how companies are evaluating their large language models. openai has evals, which is “a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.” anthropic has written about "constituti...
Join discussionAug 23, 2025 · 4 min read · disclaimer: while i am a huge notion fan, i am not an employee (well, for now, anyway 😏), so this is just my best guess at what’s going on under the hood. still, as someone who geeks out over infra, i couldn’t resist talking about why this particula...
Join discussion