LangGraph Foundations: Graph-Based LLM Prompting and Chaining Training Course

Introduction to LangGraph and Graph-Theoretic Concepts

• The rationale for using graphs in LLM apps: orchestration benefits over simple linear chains
• Understanding nodes, edges, and state within LangGraph
• Getting started: building the first executable graph

State Management and Prompt Sequencing

• Structuring prompts as discrete graph nodes
• Managing data flow between nodes and processing outputs
• Memory strategies: distinguishing short-term context from persistent storage

Branching Logic, Control Flow, and Resilience

• Implementing conditional routing and multi-path execution strategies
• Handling retries, timeouts, and defining fallback mechanisms
• Ensuring idempotency and safe re-execution of processes

Tool Usage and External System Integration

• Executing function and tool calls directly from graph nodes
• Interacting with REST APIs and services integrated within the graph
• Processing and utilizing structured output formats

Retrieval-Augmented Generation (RAG) Workflows

• Basics of document ingestion and text chunking
• Working with embeddings and vector databases (e.g., ChromaDB)
• Generating grounded responses with accurate citations

Validation, Troubleshooting, and Performance Assessment

• Writing unit-level tests for specific nodes and execution paths
• Leveraging tracing and observability tools for insights
• Quality assurance: verifying factuality, safety, and deterministic behavior

Deployment Strategies and Packaging Basics

• Setting up environments and managing project dependencies
• Exposing graph workflows via API endpoints
• Managing workflow versions and implementing rolling updates

Recap and Future Directions