Spec / 010 · curriculum

Learn

The curriculum is eleven lessons: foundations through reliability, then coding agents. Start with how models work, then walk the stack until you can eval and ship. The 3D labs are lessons, not decoration.

1. 01

   Open the curriculum

2. 02

   LLM foundations

3. 03

   Watch the web research agent

4. 04

   Reliability engineering

5. 05

   Vibe coding pipeline

6. 06

   Prompt workshop

7. 07

   Blog notes

All guides

• LLM Foundations: History, Providers, Context, and Reasoning
• Prompt Engineering: System Prompts, Examples, and Control
• Agent Fundamentals: Autonomy, the Loop, Structured Output, Routing
• Tool-Using Agents: Calling, Design, Memory, Middleware
• MCP and Integrations: Servers, Clients, and Third-Party Tools
• Agentic Workflows: Branch, Chain, Merge, Suspend, Stream
• RAG and Knowledge: Chunking, Embeddings, Rerank, Alternatives
• Multi-Agent Systems: Supervisors, Subagents, Parallel, A2A
• Production Engineering: Tracing, Evals, Local Dev, Deploy
• Agent Reliability Engineering: Safe to Run 10,000 Times
• Advanced Agents: Coding, Sandboxes, Filesystems, Multimodal
• Agent Architecture
• Vibe Coding

From the blog

• Watch the Agent, Then Read the Prompt
• What Is Vibe Coding? A Practical Definition for 2026
• 7 Prompt Engineering Mistakes Beginners Make (and How to Fix Them)
• Multi-Agent Systems Explained: When One AI Agent Is Not Enough
• Building Your First AI Agent: A Step-by-Step Guide