Become an AI engineer

# Agents and harnesses

Agents and harnesses is a unit of 13 lessons in Rabbithole's Become an AI engineer course, at levels 3 to 5.

## Lessons

1. Getting started with AI agents · Level 3
2. Agents: what they are and how you build and update them · Level 3
3. Tool use: who runs the tool, and how to read it · Level 3
4. Workflow Automation: Triggers, Steps and Checks · Level 3
5. Agent memory: letting later runs learn from earlier ones · Level 3
6. Agent Skills · Level 3
7. Agent workflow patterns · Level 3
8. Computer use: the model acts, you return a screenshot · Level 3
9. Multi-Agent Systems · Level 3
10. Sandboxing and Execution Environments · Level 3
11. Voice Agents: Speech In, Speech Out · Level 3
12. Formal verification: proving software correct · Level 4
13. Automated Theorem Proving with LLM Agents: COPRA · Level 5

## Sources

These lessons draw on 18 sources. The most used:

- [OpenAI Agents SDK (Python)](https://github.com/openai/openai-agents-python)
- [Claude Docs — Tool Use and Agent Tools](https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview)
- [Claude Docs — Managed Agents](https://platform.claude.com/docs/en/managed-agents/overview)
- [Advanced LLM Agents MOOC, Spring 2025](https://www.youtube.com/playlist?list=PLS01nW3RtgorL3AW8REU9nGkzhvtn6Egn)
- [OpenAI Docs — Agents and Agents SDK](https://developers.openai.com/api/docs/guides/agents/sdk)
- [Claude Quickstarts](https://github.com/anthropics/claude-quickstarts)
- [Hugging Face Agents Course](https://github.com/huggingface/agents-course)
- [Cloudflare Workers AI, Agents, AI Gateway and Vectorize](https://developers.cloudflare.com/workers-ai/)

[All units in Become an AI engineer](https://rabbithole.pro/courses/ai-engineer)

Source page: https://rabbithole.pro/courses/ai-engineer/agents-and-harnesses