Foundation Model Agent Loop

A plain Python agentic loop that calls a Databricks-hosted Foundation Model with tool definitions. The model decides which tools to call; your code executes them and feeds results back until the model produces a final text response.

No LangGraph, no MLflow, no custom model deployment — just HTTP calls.

When to Use This Skill

Use this skill when:

• Building a backend agent that uses a Databricks Foundation Model (e.g., databricks-meta-llama-3-3-70b-instruct) as the "brain"
• The agent needs to call tools (Genie, business logic, external APIs) based on user input
• You want the model to decide routing — no hand-written if/else classifiers

Prerequisites