Build applications that make decisions (chatbots, agents, simulations, etc...). Monitor, trace, persist, and execute on your own infrastructure.
Python aiburrchatbot-frameworkdagsgenerative-aigraphshacktoberfestllmopsllmsmlopspersistent-data-structurestate-machinestate-managementvisibility
Tool-calling
This example shows the basic tool-calling design pattern for agents.
While this leverages the OpenAI API, the lessons are the same whether you use different tool-calling APIs (E.G. Anthropic), or general structured outputs (E.G. with instructor).
Rather than explain the code here, we direct you to the blog post
Files
- application.py -- contains code for calling tools + orchestrating them
- notebook.ipynb -- walks you through the example with the same code
- [requirements.txt] -- install this to get the right environment