Loading...
A 12-week curriculum covering the full agent stack — the ReAct loop, planning, memory, tool use, multi-agent frameworks, computer-use and voice agents, prompt-injection defense, and eval-driven development.
LLM Study Plan
Familiarity with the LLM study plan is recommended →
Weeks 1-4
The agent loop, decomposition strategies, and tool use
Weeks 5-8
Long-term memory and the major orchestration frameworks
Weeks 9-12
Specialized agents, safety, observability, and benchmarks
What an agent is, the perceive-think-act loop, the ReAct pattern, what early agents got wrong, and the modern agent design space.
Decoupling planning from execution — Plan-and-Execute, ReWOO's DAG planner, hierarchical task networks, and when to plan up-front vs react.
Verbal reinforcement learning, self-critique loops, tree-structured reasoning, and Language Agent Tree Search.
Function calling fundamentals, JSON-mode and tool schemas, parallel tool calls, the Model Context Protocol, and tool-result handling.
The context-window problem, MemGPT's OS-inspired paging, memory blocks, sleep-time compute, and memory promotion.
Vector + graph + lexical hybrid retrieval, the Mem0 architecture, Voyager-style skill libraries, and how to evaluate memory quality.
Anthropic's workflow taxonomy — chaining, routing, parallelization, orchestrator-worker — and LangGraph's stateful-graph runtime with checkpointing.
AutoGen's actor model, CrewAI roles, OpenAI Agents SDK, Claude Agent SDK with subagents, the Agno and Mastra runtimes, and a framework selection matrix.
Computer-use agents that drive a screen, action grammars and grounding, Pipecat voice pipelines, LiveKit streaming, and latency budgets for realtime.
Multi-agent debate and consensus, chairperson moderation, the failure taxonomy of agentic systems, cascading tool failures, and orchestration patterns for coordination.
The prompt-injection taxonomy, direct vs indirect attacks, data tagging and trust boundaries, output validation firewalls, and red-teaming agents.
SWE-bench, GAIA, WebArena, and OSWorld benchmarks, OpenTelemetry GenAI conventions, observability platforms, production runtimes, and the eval-driven development loop.
Curriculum designed to take you from the simplest ReAct loop to deploying coordinated, observable, hardened multi-agent systems in production.