The flagship agent-building course. You will pin down what separates an agent from a workflow, design tools agents can actually use, connect them to real systems over MCP, engineer context deliberately, add short and long term memory, orchestrate multi-agent systems with human-in-the-loop gates, run computer-use and browser agents, and trace and debug agent runs (evaluation harnesses live in the AI Evaluation and Reliability course). The capstone takes one agent from prototype to production.
Prerequisites: Building Your First AI Product
2 Modules · 9 Lessons