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

Introduction to LangGraph and Graph Theories

• The rationale for using graphs in LLM applications: orchestrating complexity versus simple chains
• Understanding nodes, edges, and state within LangGraph
• Getting started: executing your first LangGraph workflow

State Management and Prompt Chaining

• Structuring prompts as discrete graph nodes
• Managing state transitions between nodes and processing outputs
• Memory strategies: differentiating between short-term and persisted context

Branching, Control Flow, and Error Handling

• Implementing conditional routing and multi-path workflow designs
• Managing retries, timeouts, and defining fallback strategies
• Ensuring idempotency and facilitating safe re-executions

Tools and External Integrations

• Executing function and tool calls directly from graph nodes
• Interacting with REST APIs and services within the graph structure
• Handling and processing structured data outputs

Retrieval-Augmented Workflows

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

Testing, Debugging, and Evaluation

• Writing unit-level tests for individual nodes and execution paths
• Implementing tracing and observability mechanisms
• Conducting quality assessments for factuality, safety, and determinism

Packaging and Deployment Essentials

• Setting up development environments and managing dependencies
• Exposing graph workflows via API endpoints
• Versioning workflows and managing rolling updates

Course Wrap-Up and Future Directions