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 Theories
- The rationale for using graphs in LLM applications: the advantages of orchestration over simple sequential chains.
- Defining the roles of nodes, edges, and state within the LangGraph ecosystem.
- Getting started: building and executing your first functional graph.
State Management and Prompt Sequencing
- Architecting prompts as discrete graph nodes.
- Managing the flow of state between nodes and processing resulting outputs.
- Implementing memory strategies: distinguishing between short-term context and persisted data.
Branching, Control Logic, and Error Resolution
- Configuring conditional routing to support multi-path workflows.
- Establishing robust error handling through retries, timeouts, and fallback mechanisms.
- Ensuring system stability via idempotency and safe re-execution protocols.
Tool Usage and External Integration
- Executing function and tool calls directly from graph nodes.
- Interacting with REST APIs and external services within the graph structure.
- Handling and utilizing structured data outputs effectively.
Retrieval-Augmented Generation (RAG) Workflows
- Foundations of document ingestion and text chunking.
- Leveraging embeddings and vector databases, such as ChromaDB.
- Generating grounded responses with accurate source citations.
Testing, Troubleshooting, and Performance Evaluation
- Developing unit-level tests for specific nodes and execution paths.
- Implementing tracing mechanisms and observability tools.
- Conducting quality assurance checks focusing on factuality, safety, and deterministic behavior.
Deployment and Packaging Essentials
- Managing development environments and dependencies.
- Exposing graph applications via API endpoints.
- Managing workflow versioning and implementing rolling updates.
Conclusion and Future Directions
Requirements
- A solid grasp of fundamental Python programming principles.
- Practical experience with REST APIs or Command Line Interface (CLI) tools.
- Basic knowledge of LLM principles and core prompt engineering techniques.
Target Audience
- Developers and software engineers new to the orchestration of graph-based LLMs.
- Prompt engineers and emerging AI practitioners focused on building complex, multi-step LLM applications.
- Data professionals investigating the automation of workflows using Large Language Models.