Module 1 · Building AI Agents: Harness Engineering
AI Agent Fundamentals: lessons
What an agent is, what a harness is, and when a problem actually needs an agent — and when it does not.
4 lessons · 1h 13m · 1 free to read now
What this module teaches, what it asks of you first and how you practise it are on its syllabus page.
Open the syllabusAn agent is an LLM run in a loop with tools. The model proposes actions; the harness decides what actually happens.
Designing Tools for Agents
16 minNames, descriptions and schemas that get a model to use a tool correctly.
Included with the track Simulation insideWorkflows versus Agents
14 minPrompt chaining, routing, parallelisation — patterns simpler than a full agent, and when each one is enough.
Included with the track Simulation insideBuild an Agent from Scratch
25 minThe loop, the tools and the stop conditions in plain Python — tested against a scripted model first, then run with the Anthropic SDK.
Included with the track Simulation inside