A structured learning path for software engineers entering the AI agent ecosystem. Assumes programming knowledge. Python examples throughout.
By the end of this guide you will be able to:
- Distinguish between LLMs, agents, and multi-agent systems
- Build agents that reason, use tools, and maintain memory
- Implement RAG pipelines for knowledge-augmented agents
- Evaluate, observe, and deploy agents in production
| # | Module | Topics |
|---|---|---|
| 00 | Foundations | LLM vs Agent, LLM fundamentals, prompting, safety |
| 01 | Agent Architecture | Agent loop, tools, memory, planning, multi-agent |
| 02 | Building Agents | First agent, RAG, structured output, frameworks |
| 03 | Production | Evaluation, observability, cost optimization, deployment |
| 04 | Workshop | Full agent: web search, document reading, citations |
- Read modules in order. Each builds on the previous.
- Every code example is runnable. Copy, modify, break, fix.
- Refer to AGENTS.md for the design principles behind this guide.
- 2+ years software engineering experience
- Python proficiency
- Familiarity with APIs, state management, system architecture
- No prior AI/ML experience required