Apr 13 · 11 min read · I can tell, within about two minutes of joining a new team, whether they're serious about building with LLMs. I don't have to look at the code. I don't have to see the prompts. I ask one question: "Show me your eval." A team that's serious says "here...
Join discussion
Nov 16, 2025 · 14 min read · You want to build an AI application. Something useful. Something your users can rely on. But here's the problem: How do you know it's actually good? You can build an app that generates summaries, answers questions, or writes emails. It works. Your te...
Join discussion
Oct 26, 2025 · 5 min read · A few months ago, as I was exploring machine learning while working on a project, one of my models kept behaving in a weird way. I had built a classifier to detect cats in images. During training, accuracy was awesome, near 99%. But in production, it...
Join discussion
Oct 6, 2025 · 7 min read · I recently came across a paper with a title that neatly sums up the state of AI: "What the F*ck Is Artificial General Intelligence?" by Michael Timothy Bennett. It's direct, unlike so much AI slop out there, and it asks exactly the right question. Bu...
Join discussion
Sep 19, 2025 · 6 min read · When we started rolling out voice and chat agents at Hillflare, it felt like opening a thousand tabs at once. Every new client, every script, every accent, every “quick tweak” to a prompt multiplied the ways things could go right… or sideways. Readin...
Join discussion
Jul 10, 2025 · 8 min read · Introduction In AI development, evaluating an LLM’s performance using test cases based on your understanding of what the LLM is supposed to do is critical. These evaluations, commonly called “Evals”, serve as test cases to help you assess whether you...
HHaniel commented
Jul 4, 2025 · 2 min read · Увидев в списке “примитивов” (см. предыдущий пост) … , конечно, захотелось посмотреть на зависимости eval, мимо которого пройти просто невозможно. Во-первых, код определения. (NIL "eval" _Eval) src/glob.l # (eval 'any ['cnt]) -> any (de _Eval (Ex...
Join discussion
