Build AI agents from scratch. Learn the foundational patterns—ReAct, tool calling, multi-agent orchestration—before diving into the Tenzai codebase.
Create a minimal agent using raw API calls. Understand the core loop before using frameworks.
Refactor your agent using Pydantic AI for structured outputs, validation, and cleaner tool definitions.
Build a stateful agent with LangGraph. Learn checkpointing, branching, and human-in-the-loop patterns.
Coordinate multiple specialized agents. Build a supervisor pattern for complex task decomposition.
Build identical agents in three frameworks. Use HTTP proxy to discover implementation differences.