🧠 Learning track · 30 free lessons
Advanced AI Engineering
Engineer-grade depth: how LLMs work, evals, agents, and production AI.
LLM internals, evals, agents, LLM Council, /loop, autoresearch — engineer-grade depth. Free forever · no sign-up to start · English & Spanish.
What you'll do
- Understand how LLMs actually work — tokens, sampling, context windows.
- Build eval suites, tool-using agents, multi-agent systems and the LLM Council pattern.
- Learn the /loop and autoresearch patterns for real agent work.
- Ship it: observability, cost engineering, security, and local models.
All 30 lessons in this track
- 71 How LLMs actually work
- 72 Tokens and context windows
- 73 Temperature and sampling
- 74 Choosing the right model
- 75 Advanced prompting techniques
- 76 Evals 101
- 77 Building eval suites
- 78 Fine-tuning concepts
- 79 RAG vs fine-tuning vs long context
- 80 Checkpoint: advanced RAG
- 81 Agents 101
- 82 Tool-using agents
- 83 The /loop pattern
- 84 Multi-agent systems
- 85 The LLM Council pattern
- 86 The autoresearch pattern
- 87 Agent orchestration in practice
- 88 Agent memory
- 89 Agent safety and guardrails
- 90 Checkpoint: build a research agent
- 91 Production architecture
- 92 Observability and tracing
- 93 Scaling and cost engineering
- 94 Security for AI systems
- 95 Responsible AI in practice
- 96 Staying current
- 97 Open-source and local models
- 98 The AI engineer role
- 99 Portfolio and career
- 100 Final capstone: your AI product
Free printable cheat sheets
Explore the other tracks