Teaching · Build AI Agents

A curriculum for building AI agents in production

Eight modules from transformer foundations to agent operations. Code, runnable demos, and the live agent live in the build-ai-agents repository. These pages mirror the same source of truth, rendered for reading.

3 / 8 live10 lessonspinned a4648b1
Module 01
Transformer FoundationsStrong basics are non-negotiable before building agents.
3 lessons
Module 02
RAG (Retrieval-Augmented Generation)The first enterprise use case of LLMs since 2022. Immediately useful at work.
3 lessons
Module 03
Agentic WorkflowsRigid workflows with some agentic control + MCP integration.
4 lessons
Module 04
Full Agents (Human-in-the-Loop)Agent has full control plus human oversight and hooks for safety.
  • Agent autonomy and the autonomy ladder
  • Human-in-the-loop interrupts and approvals
  • Safety hooks and policy enforcement
In progress
Module 05
Skills and SandboxesThe next 3-6 months of agent evolution.
  • The Claude Skills pattern (capabilities packaged as files)
  • Sandboxed code execution for agents
  • When 'agent skills' is just rebranded prompt + tools
In progress
Module 06
Coding AgentsAGENTS.md plus tools plus repo structure for AI-first development.
  • The CLAUDE.md / AGENTS.md pattern for instructing agents
  • Hooks, permissions, environment configuration
  • Iteration planning across long-running coding tasks
In progress
Module 07
Personal AgentsAlways-on agents that live in the messengers you already use.
  • Daemon agents vs per-invocation CLIs (OpenClaw vs Claude Code)
  • Telegram and WhatsApp gateway architecture
  • Kubernetes deployment of personal agents
In progress
Module 08
Agent OperationsProduction-grade observability, security, cost, and reliability for agentic systems.
  • Tracing and observability — span design for agentic loops
  • Cost engineering — token accounting, prompt caching ROI
  • Production evals and failure forensics
Coming soon