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

Introduction to LangGraph and Graph-Theoretic Concepts

• The rationale for using graphs in LLM applications: contrasting orchestration with simple linear chains
• Defining and understanding nodes, edges, and state within the LangGraph ecosystem
• Getting started: executing the first runnable graph

State Management and Advanced Prompt Chaining

• Structuring prompts as functional graph nodes
• Managing data flow by passing state between nodes and processing outputs
• Implementing memory patterns: distinguishing between short-term context and persisted data

Branching Logic, Control Flow, and Robust Error Handling

• Implementing conditional routing and managing multi-path workflow executions
• Configuring retry mechanisms, timeout thresholds, and fallback strategies
• Ensuring idempotency and executing safe re-runs

Tool Utilization and External System Integrations

• Executing function and tool calls directly from graph nodes
• Invoking REST APIs and external services within the graph structure
• Processing and utilizing structured output formats

Retrieval-Augmented Generation (RAG) Workflows

• Foundations of document ingestion and text chunking
• Utilizing embeddings and vector databases (such as ChromaDB)
• Generating grounded responses with accurate source citations

Quality Assurance: Testing, Debugging, and Evaluation

• Developing unit-style tests for individual nodes and complete workflow paths
• Implementing tracing mechanisms and enhancing observability
• Performing quality checks focused on factuality, safety standards, and deterministic behavior

Deployment Strategies and Packaging Basics

• Configuring development environments and managing dependencies
• Serving graph workflows via API endpoints
• Implementing workflow versioning and managing rolling updates

Course Summary and Recommended Next Steps