What are the principles we can use to build LLM-powered software that is actually good enough to put in the hands of production customers?
Building Agents with LLM structured generation (BAML), MCP Tools, and 12-Factor Agents principles
Tiny Agent: MCP Agent that think in Python
Tiny Agent: MCP Agent that think in Python
🐣 TinyAgent that thinks in python and execute it in the cloud with extendable hooks