AI agent development related content
ai-agent
Aug 21, 2026
•
9 min read
You already know how to build this. Here is what changes when the parts are AI agents.
Aug 15, 2026
4 min read
Every AI agent you've ever used is a while-loop. Most of its bugs live in how that loop stops.
Aug 8, 2026
6 min read
You didn't build an agent. You built a harness and pointed it at a model.
Jul 26, 2026
5 min read
Your agent sends the same 20,000 tokens every turn. Your provider will re-sell them at a tenth of the price — if your prompt is built for it.
Jul 19, 2026
One model for everything feels simple. It's actually the most expensive routing decision you never made.
Jul 11, 2026
Every MCP server you connect makes your agent more capable on paper — and worse in production. Here's the failure, and the ~50 lines of Python that fix it.
Jul 3, 2026
A crash tutorial. Working traces before lunch. Production-hardened by end of week.
Jun 26, 2026
How agent observability differs from the distributed tracing you already know — and exactly what to build in your first week.
Jun 19, 2026
11 min read
Why fault tolerance for agents is a different problem than fault tolerance for microservices — and what to build first.
Jun 12, 2026
The cost math nobody warns you about — and how to find out if you have the problem tonight.
Jun 5, 2026
What a circuit breaker actually does, why multi-agent systems fail without one, and how to build one this week.
Apr 25, 2026
3 min read
PLUS: How to stop Hoarding Tiny Tasks
Apr 7, 2026
PLUS: Eat the Frog — Kill Your Hardest Task First
