Before writing any LLM logic, define your evaluation step. Here's how evals catch bad outputs early on production systems.
Ship AI Features Without the Fire Drill: Write the Eval First
Related
Agent Memory Is One Forgotten Filter Away From a Leak
AI agents are getting long-term memory in shared tables. I think the boundary between customers belongs in PostgreSQL, and I'm looking for teams who want theirs tested.
Agent Calling Agent Tools (Part 3): Integrating gemini-api CLI & Keyless ADC Skills inside Antigravity CLI (agy)
Verified baseline: 2026-11-14 22:07 (Asia/Taipei, UTC+8), gemini-api 0.4.6, upstream snapshot...
Our Slack AI agent can read our code. It can't ship it.
How I built a guardrail that lets an AI agent reason over private source code without ever posting it to Slack: nine exits, five control layers, code fingerprints, and a log-only w...