Learning Tracks
Your roadmap to real skills.
Many tracks and dozens of sub-tracks. Every track starts with a free intro lesson.
Learning Tracks
Many tracks and dozens of sub-tracks. Every track starts with a free intro lesson.
Build agents from scratch in Python with full control.
~13 hours · 11 lessons
View track details →Code-First Agent Fundamentals: Python, ReAct, and Tool Calling
Set up a Python environment, understand the ReAct loop, and implement a simple tool-calling agent from scratch using the Anthropic or OpenAI API.
LangChain: The Agent Framework
Build agents, chains, and memory-backed conversations with LangChain: connect tools, define custom agents, and manage state across multi-turn interactions.
LangGraph: Stateful Multi-Step Agent Graphs
Use LangGraph to build graph-based agentic workflows with conditional branching, cyclic reasoning loops, and persistent checkpointing for long-running tasks.
LlamaIndex: RAG Pipelines and Data Agents
Index documents in a vector store, build a retrieval-augmented generation pipeline, and create a data agent that can query multiple sources and synthesise answers.
CrewAI: Multi-Agent Teams with Roles and Tasks
Define specialised agent roles (researcher, writer, reviewer), assign tasks, and orchestrate a crew of agents that collaborate to complete complex workflows.
Raw SDK: Anthropic and OpenAI APIs from Scratch
Call AI APIs directly: implement tool use, streaming, function calling, and structured output without any framework, understanding every layer of the stack.
Model Context Protocol (MCP): Connect Agents to External Tools
Understand the Model Context Protocol, build an MCP server that exposes tools to an agent, and connect Claude or another LLM to real external systems via MCP.
Multi-Agent Systems: Orchestration, Supervisor Patterns, and When Not To
Reason about supervisor / pipeline / swarm patterns and decide honestly when a single agent beats a team.
Code-First Agent Capstone: Build and Deploy a Production Agent
Design, build, evaluate, and deploy a production-grade AI agent. Add observability, error handling, cost tracking, and a public interface.
Passion Project: Research / Security / Code-Review Agent
Pick one of three project briefs and ship a production-grade agent end-to-end with deploy URL, evals, and case study.
Code-First Agent Job Readiness
Translate code-first agent skills into a resume, portfolio plan, and interview prep for AI / agent engineer roles.