🤖 AI Agents Language
7 exercise sets. Master the precise vocabulary of autonomous AI systems — from the agent loop and tool calling to multi-agent orchestration and safety guardrails.
Agent Loop & ReAct Pattern
Master the core loop that drives every autonomous agent: think → act → observe. ReAct pattern, agent steps, trajectories, and scratchpad vocabulary.
Tool Use & Function Calling
Tool schemas, tool calls vs. tool results, tool registries, and how agents plan with tools. The vocabulary of LLM function-calling APIs.
Agent Memory
In-context vs. external memory, episodic/semantic/procedural memory types, RAG, context window pressure, and memory summarisation strategies.
Multi-Agent Orchestration
Orchestrator agents, sub-agents, crews, agent handoff, and delegation. Vocabulary for designing and discussing systems where agents collaborate.
Agent Observability
Traces, spans, token budgets, LangSmith/LangFuse vocabulary, and agent run logs. How to monitor and debug autonomous agent systems.
Agent Guardrails & Safety
Input guardrails, output guardrails, human-in-the-loop (HITL), halt conditions, and runaway tool call prevention. Safety vocabulary for agentic systems.
Agentic Architecture Patterns
Planner-executor pattern, reflection loop, verification step, memory architectures (episodic, semantic, procedural), and agent evaluation vocabulary.