Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 14 hours
Course Outline
Introduction to LangGraph and Graph Principles
- The rationale for using graphs in LLM apps: orchestration versus simple chains
- Understanding nodes, edges, and state within LangGraph
- Getting started: executing the first LangGraph graph
State Management and Prompt Chaining
- Structuring prompts as individual graph nodes
- Transferring state between nodes and processing outputs
- Memory strategies: contrasting short-term and persistent context
Branching, Control Flow, and Error Management
- Implementing conditional routing and multi-path workflows
- Managing retries, timeouts, and fallback mechanisms
- Ensuring idempotency and safe re-execution
Tools and External Integrations
- Executing function or tool calls from graph nodes
- Interacting with REST APIs and services within the graph structure
- Handling structured data outputs
Retrieval-Augmented Workflows
- Basics of document ingestion and chunking
- Utilizing embeddings and vector stores (e.g., ChromaDB)
- Generating grounded answers with citations
Testing, Debugging, and Assessment
- Writing unit-style tests for nodes and execution paths
- Implementing tracing and observability features
- Quality assurance: verifying factuality, safety, and determinism
Packaging and Deployment Essentials
- Configuring environments and managing dependencies
- Exposing graphs via API services
- Versioning workflows and managing rolling updates
Recap and Future Directions
Requirements
- A grasp of basic Python programming
- Hands-on experience with REST APIs or command-line interface tools
- Knowledge of LLM principles and foundational prompt engineering practices
Target Audience
- Developers and software engineers new to graph-based LLM orchestration
- Prompt engineers and AI newcomers constructing multi-step LLM applications
- Data practitioners investigating workflow automation using LLMs