One agent. Five days. Progressively more powerful. Learn the agent loop, tool calling, memory management, multi-agent coordination, and production deployment with real Python code in every session.
This is a text-first course that links out to the best supporting material on the internet instead of trying to replace it. The goal is to make this the best course on ai agents you can find — even without producing a single minute of custom video.
Bo runs AI agents in production for U.S. federal agencies. Every code snippet in this course comes from real systems — not toy demos built for a tutorial.
Instead of five disconnected examples, this course builds one agent that gets more capable each day. By Day 5 it is a multi-agent production system.
Instead of building framework abstractions, this course shows you the raw loop and links to LangChain, CrewAI, and LlamaIndex so you understand what they're abstracting.
Each day finishes in about an hour of focused reading plus hands-on coding. No live classes, no quizzes, no calendar commitment.
Each day stands alone. Read them in order for the full picture, or jump straight to the day that answers the question you have today.
The agent loop: perceive, reason, act. How tool calling works under the hood. Why agents fail and what makes them succeed. Your first working agent in Python.
JSON schema tool definitions. Web search, file read/write, code execution, API calls. Tool selection logic. Building a 5-tool agent that can actually accomplish tasks.
Conversation history management. Token budget. Vector store memory with embeddings. The difference between short-term and long-term memory and when you need each.
Orchestrator-subagent patterns. How to split a complex task across specialized agents. Shared memory, message passing, and preventing agent conflicts.
Error handling, retry logic, cost controls, budget caps, logging, monitoring. FastAPI wrapper to serve your agent as an API. The checklist before you ship.
Instead of shooting our own videos, we link to the best deep-dives already on YouTube. Watch them alongside the course. All external, all free, all from builders who ship this stuff.
Conceptual deep-dives on what LLM agents are, how the perception-reasoning-action loop works, and what distinguishes good agent architectures.
Practical walkthroughs of Claude's tool calling API with real code examples you can run immediately.
LangChain agent tutorials showing the framework abstractions built on top of the raw tool-calling patterns this course covers.
Deep dives on orchestrator-subagent architectures and multi-agent frameworks like CrewAI and AutoGen.
How retrieval-augmented generation works — the foundation of long-term agent memory using embedding models and vector stores.
FastAPI-based deployment patterns for AI agents — serving, auth, rate limiting, monitoring in production systems.
The best way to go deeper on any topic is to read canonical open-source implementations. These repositories implement the core patterns covered in this course.
The most widely-used agent framework. Read the agents/ directory to understand how the framework wraps raw tool calling.
Multi-agent orchestration with role-based agents and task delegation. Clean implementation of the orchestrator-subagent pattern.
Microsoft's multi-agent conversation framework — a different coordination model where agents communicate via conversation.
The canonical library for RAG and agent memory. Best reference for vector store retrieval integrated into agent context management.
You know Python and want to build agents that actually work in production. This course skips toy examples and goes straight to production patterns.
You need to pick between LangChain, CrewAI, AutoGen, and raw API calls. This course explains underlying patterns so you can evaluate frameworks from first principles.
You are responsible for an AI roadmap and need to understand what agents can and can't do, what they cost, and what reliability actually requires.
The 2-day in-person Precision AI Academy bootcamp covers real agent engineering — tools, loops, memory, deployment — hands-on with Bo. 5 U.S. cities. $1,490. 40 seats max. June–October 2026 (Thu–Fri).
Reserve Your Seat