5 min read
LLM Structured Outputs Always Parse Now. They Can Still Be Wrong.
Schema enforcement killed the parse errors. It didn't make the answers true, and a hallucinated ID passes every type check.
13 articles in AI Engineering.
5 min read
Schema enforcement killed the parse errors. It didn't make the answers true, and a hallucinated ID passes every type check.
7 min read
Your agent has tests. The thing grading them doesn't. Five rules for a judge you can trust.
8 min read
The 4 observability decisions to make before launch: trace every step, one ID end to end, five numbers on a dashboard, and logs that link to the trace.
6 min read
The MCP servers I actually reach for when building features and shipping production apps.
8 min read
Checkpoints, scoped retries, idempotent tools, and run budgets: how production agents survive the failure math of 20-step loops.
9 min read
Your agent isn't a user. Stop authenticating it like one.
8 min read
Mock the model and your suite goes green while testing nothing. Here's what to run instead, from unit tests to live alerts.
9 min read
One good agent beats five mediocre ones. Here's the pressure that tells you when the split is finally worth it.
9 min read
7 rules for tool definitions a model can't misread, starting with the sentence that fixes the most bugs.
9 min read
The demo works on your machine. Here's what takes it down when real users show up.
7 min read
Learn how to build software smartly with Advisor, Generator & Reviewer AI Agents
7 min read
A practical guide to maintain quality at AI speed
5 min read
Learn how to improve your IDE and increase your coding speed with these extensions.