Course Outline

LangGraph and Agent Patterns: A Practical Primer for Government

  • Comparing Graphs to Linear Chains: When and Why for Government Operations
  • Agents, Tools, and Planner-Executor Loops in Public Sector Applications
  • Hello Workflow: A Minimal Agentic Graph for Government Use

State, Memory, and Context Passing for Government Systems

  • Designing Graph State and Node Interfaces for Government Operations
  • Short-Term Memory vs. Persisted Memory in Government Applications
  • Context Windows, Summarization, and Rehydration for Government Workflows

Branching Logic and Control Flow for Government Processes

  • Conditional Routing and Multi-Path Decisions in Government Systems
  • Retries, Timeouts, and Circuit Breakers for Robust Government Operations
  • Fallbacks, Dead-Ends, and Recovery Nodes in Government Workflows

Tool Use and External Integrations for Government Applications

  • Function/Tool Calling from Nodes and Agents in Government Systems
  • Consuming REST APIs and Databases from the Graph for Government Operations
  • Structured Output Parsing and Validation for Government Workflows

Retrieval-Augmented Agent Workflows for Government

  • Document Ingestion and Chunking Strategies for Government Data
  • Embeddings and Vector Stores with ChromaDB for Government Use
  • Grounded Responses with Citations and Safeguards for Government Applications

Evaluation, Debugging, and Observability for Government Systems

  • Tracing Paths and Inspecting Node Interactions in Government Workflows
  • Golden Sets, Evaluations, and Regression Tests for Government Operations
  • Quality, Safety, and Cost/Latency Monitoring for Government Applications

Packaging and Delivery for Government Use

  • FastAPI Serving and Dependency Management for Government Systems
  • Versioning Graphs and Rollback Strategies in Government Operations
  • Operational Playbooks and Incident Response for Government Workflows

Summary and Next Steps for Government Applications

Requirements

  • Demonstrated proficiency in Python
  • Experience developing large language model (LLM) applications or prompt chains
  • Familiarity with REST APIs and JSON data formats

Audience for Government

  • AI engineers
  • Product managers
  • Developers working on interactive LLM-driven systems
 14 Hours

Number of participants


Price per participant

Upcoming Courses

Related Categories