Extract the Fact, Judge in Code
Aug 15, 2026 · AI agents · 4 min read — Ask a model whether the budget is big enough and it decides, quietly, what big enough means.
Blog
Notes From The Build
Short, honest notes on AI agents, automation, and the craft between builds.
Field notes from building AI agents for real teams — the prompts that held up, the wiring that didn't, and the judgment calls in between. I'm Kershey Cariño; each note is a principle you can lift straight into your own build.
The notes
Structured Is Not Constrained
Aug 15, 2026 · AI agents · 4 min read — Structured output guarantees your code can read the answer, never how many answers were possible.
Cache the Thing, Not the Prompt
Aug 15, 2026 · AI agents · 4 min read — The reason your automation pays twice for the same answer is the key, not the cache.
Feed Your AI Only What It Needs
Jul 5, 2026 · AI agents · 3 min read — The more work you hand a language model, the less you can trust any single thing it gives back.
Why a Good AI Assistant Says "I Don't Know"
Jul 5, 2026 · AI agents · 2 min read — The one refusal that turns a clever bot into a teammate people trust.
The craft behind these notes
The tools and skills these builds lean on: discover my mastery.