Track
Agents & Tooling
Goal: Tool use, ReAct, memory, and browser/computer-use agents.
Prereqs: LLMs.
Status: done
Work through the steps in order. Bold links open YouTube.
| Step | Concept | YouTube | Read |
|---|---|---|---|
| 1 | What an agent loop actually is | Anthropic — Building Effective AI Agents | |
| 2 | ReAct: reasoning and acting | Yao et al. 2022 — ReAct: Synergizing Reasoning and Acting in Language Models | |
| 3 | Build a ReAct agent yourself, no framework | Python: Create a ReAct Agent from Scratch | |
| 4 | Tool use and function calling | James Briggs — Building Chatbot Agents from Scratch with OpenAI Functions | Schick et al. 2023 — Toolformer |
| 5 | Long-term memory for agents | Packer et al. 2023 — MemGPT: Towards LLMs as Operating Systems | |
| 6 | Browser agents | Build an AI Agent that Browses the Web | browser-use — make websites accessible to AI agents |
| 7 | Computer-use agents | Claude — Computer use for coding | Claude Docs — Computer use tool |
Found a broken link or an unclear step? Report a problem with this track.