Get in Touch
 Duration 14 hours

Course Outline

Overview of LangGraph and Graph Architecture

  • The case for graph structures in LLM applications: orchestration versus linear chains
  • Understanding nodes, edges, and state within LangGraph
  • Initial setup: Creating the first executable graph

State Handling and Prompt Sequencing

  • Structuring prompts as individual graph nodes
  • Managing state transfer between nodes and processing outputs
  • Implementing memory strategies: distinguishing between short-term and persistent context

Logical Branching, Control Flow, and Error Resilience

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

Tool Integration and External Connectivity

  • Executing function or tool calls directly from graph nodes
  • Interacting with REST APIs and external services within the graph
  • Processing and utilizing structured data outputs

Retrieval-Augmented Generation Workflows

  • Foundations of document ingestion and text chunking
  • Utilizing embeddings and vector databases (e.g., ChromaDB)
  • Generating grounded responses with source citations

Testing, Troubleshooting, and Quality Evaluation

  • Writing unit-style tests for specific nodes and execution paths
  • Implementing tracing and observability tools
  • Conducting quality assurance checks for factuality, safety, and determinism

Deployment and Packaging Basics

  • Configuring environments and managing software dependencies
  • Exposing graph applications via API endpoints
  • Managing workflow versions and implementing rolling updates

Conclusion and Future Directions

Requirements

  • Proficiency in fundamental Python programming
  • Practical experience with REST APIs or command-line interface tools
  • Working knowledge of LLM principles and basic prompt engineering

Target Audience

  • Software developers and engineers beginning their journey into graph-based LLM orchestration
  • Prompt engineers and AI specialists designing multi-step LLM applications
  • Data practitioners investigating workflow automation through LLM integration

Number of participants


Price per participant

Provisional Upcoming Courses (Require 5+ participants)

Related Categories