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
LangGraph and Agent Patterns: A Practical Introduction
- Comparing graphs versus linear chains: scenarios and rationale.
- Understanding agents, tools, and planner-executor loops.
- Building a basic agentic graph: a 'Hello Workflow' example.
State, Memory, and Context Management
- Structuring graph state and defining node interfaces.
- Distinguishing between short-term and persisted memory.
- Managing context windows, summarization, and data rehydration.
Branching Logic and Control Flow
- Implementing conditional routing and multi-path decision-making.
- Managing retries, timeouts, and circuit breaker patterns.
- Designing fallbacks, handling dead-ends, and recovery nodes.
Tool Usage and External Integrations
- Executing function and tool calls from graph nodes and agents.
- Interacting with REST APIs and databases directly from the graph.
- Parsing and validating structured outputs.
Retrieval-Augmented Agent Workflows
- Strategies for document ingestion and chunking.
- Utilizing embeddings and vector stores with ChromaDB.
- Generating grounded responses with citations and safety safeguards.
Evaluation, Debugging, and Observability
- Tracing execution paths and analyzing node interactions.
- Establishing golden sets, evaluations, and regression tests.
- Monitoring quality, safety, and cost/latency metrics.
Packaging and Deployment
- Serving via FastAPI and managing dependencies.
- Versioning graphs and implementing rollback strategies.
- Developing operational playbooks and incident response procedures.
Summary and Recommended Next Steps
Requirements
- Proficient command of Python.
- Hands-on experience in developing LLM applications or managing prompt chains.
- Understanding of REST API architecture and JSON data formats.
Target Audience
- Specialized AI Engineers.
- Strategic Product Managers.
- Developers focused on creating interactive, LLM-driven systems.