Track 01

Build AI Agents

From transformer foundations to production agent operations.

8 modules · 10 lessons published · pinned a4648b1

Transformer FoundationsStrong basics are non-negotiable before building agents.

RAG (Retrieval-Augmented Generation)The first enterprise use case of LLMs since 2022. Immediately useful at work.

Agentic WorkflowsRigid workflows with some agentic control + MCP integration.

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

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

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

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

Agent OperationsObservability, evals, 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

All the code, runnable demos, and the live agent stay in build-ai-agents. MIT-licensed, open, regenerated when the repo changes.