Agent Engineering
42 个课时
01 The Agent Loop: Observe, Think, Act
CODE QUIZ 1 OUTPUTS
✓ → 02 ReWOO and Plan-and-Execute: Decoupled Planning CODE QUIZ 1 OUTPUTS
✓ → 03 Reflexion: Verbal Reinforcement Learning CODE QUIZ 1 OUTPUTS
✓ → 04 Tree of Thoughts and LATS: Deliberate Search CODE QUIZ 1 OUTPUTS
✓ → 05 Self-Refine and CRITIC: Iterative Output Improvement CODE QUIZ 1 OUTPUTS
✓ → 06 Tool Use and Function Calling CODE QUIZ 1 OUTPUTS
✓ → 07 Memory: Virtual Context and MemGPT CODE QUIZ 1 OUTPUTS
✓ → 08 Memory Blocks and Sleep-Time Compute (Letta) CODE QUIZ 1 OUTPUTS
✓ → 09 Hybrid Memory: Vector + Graph + KV (Mem0) CODE QUIZ 1 OUTPUTS
✓ → 10 Skill Libraries and Lifelong Learning (Voyager) CODE QUIZ 1 OUTPUTS
✓ → 11 Planning with HTN and Evolutionary Search CODE QUIZ 1 OUTPUTS
✓ → 12 Anthropic's Workflow Patterns: Simple Over Complex CODE QUIZ 1 OUTPUTS
✓ → 13 LangGraph: Stateful Graphs and Durable Execution CODE QUIZ 1 OUTPUTS
✓ → 14 AutoGen v0.4: Actor Model and Agent Framework CODE QUIZ 1 OUTPUTS
✓ → 15 CrewAI: Role-Based Crews and Flows CODE QUIZ 1 OUTPUTS
✓ → 16 OpenAI Agents SDK: Handoffs, Guardrails, Tracing CODE QUIZ 1 OUTPUTS
✓ → 17 Claude Agent SDK: Subagents and Session Store CODE QUIZ 1 OUTPUTS
✓ → 18 Agno and Mastra: Production Runtimes CODE QUIZ 1 OUTPUTS
✓ → 19 Benchmarks: SWE-bench, GAIA, AgentBench CODE QUIZ 1 OUTPUTS
✓ → 20 Benchmarks: WebArena and OSWorld CODE QUIZ 1 OUTPUTS
✓ → 21 Computer Use: Claude, OpenAI CUA, Gemini CODE QUIZ 1 OUTPUTS
✓ → 22 Voice Agents: Pipecat and LiveKit CODE QUIZ 1 OUTPUTS
✓ → 23 OpenTelemetry GenAI Semantic Conventions CODE QUIZ 1 OUTPUTS
✓ → 24 Agent Observability: Langfuse, Phoenix, Opik CODE QUIZ 1 OUTPUTS
✓ → 25 Multi-Agent Debate and Collaboration CODE QUIZ 1 OUTPUTS
✓ → 26 Failure Modes: Why Agents Break CODE QUIZ 1 OUTPUTS
✓ → 27 Prompt Injection and the PVE Defense CODE QUIZ 1 OUTPUTS
✓ → 28 Orchestration Patterns: Supervisor, Swarm, Hierarchical CODE QUIZ 1 OUTPUTS
✓ → 29 Production Runtimes: Queue, Event, Cron CODE QUIZ 1 OUTPUTS
✓ → 30 Eval-Driven Agent Development CODE QUIZ 1 OUTPUTS
✓ → 31 Agent Workbench Engineering: Why Capable Models Still Fail CODE QUIZ 1 OUTPUTS
✓ → 32 The Minimal Agent Workbench CODE QUIZ 1 OUTPUTS
✓ → 33 Agent Instructions as Executable Constraints CODE QUIZ 1 OUTPUTS
✓ → 34 Repo Memory and Durable State CODE QUIZ 1 OUTPUTS
✓ → 35 Initialization Scripts for Agents CODE QUIZ 1 OUTPUTS
✓ → 36 Scope Contracts and Task Boundaries CODE QUIZ 1 OUTPUTS
✓ → 37 Runtime Feedback Loops CODE QUIZ 1 OUTPUTS
✓ → 38 Verification Gates CODE QUIZ 1 OUTPUTS
✓ → 39 Reviewer Agent: Separate Builder from Marker CODE QUIZ 1 OUTPUTS
✓ → 40 Multi-Session Handoff CODE QUIZ 1 OUTPUTS
✓ → 41 The Workbench on a Real Repo CODE QUIZ 1 OUTPUTS
✓ → 42 Capstone: Ship a Reusable Agent Workbench Pack CODE QUIZ 1 OUTPUTS
✓ →