AI Agents
9 interactive modules. Free, browser-based.
Loop + state + LLM-OS + workflows-vs-agents decision + harness anatomy.
Schemas, ACI + anti-patterns, structured outputs, MCP two ways, Skills, attack-surface preview.
When NOT to use an agent + the five named patterns + subagent topology.
Embeddings as coordinates, retrieve-then-generate, chunking, ANN trade, RAG failure modes.
Four context failure modes, four fixes, tool-loading strategy, subagents as context isolation.
Reasoning budget, ReAct + think tool, Reflexion + verifiers, when to retry, when to stop.
Compounding errors, error analysis first, golden cases, pass^k, the four eval failure modes.
Private data + untrusted content + exfiltration vector. Data-flow graphs, structural defenses, capability scoping.
RAG-only vs workflow vs autonomous agent on the same customer-order task. Compared end to end.