Building Dynamic Workflows with LangGraph and LLM Agents Training Course

LangGraph and Agent Patterns: A Practical Introduction

• Comparing graphs versus linear chains: identifying the right time and reason for each approach.
• Exploring agents, tools, and the planner-executor loop pattern.
• Building a 'Hello World' workflow: creating a minimal agentic graph.

State, Memory, and Context Management

• Defining graph state and designing node interfaces.
• Distinguishing between short-term memory and persisted storage.
• Managing context windows, implementing summarization, and rehydrating context.

Branching Logic and Control Flow

• Implementing conditional routing and multi-path decision making.
• Handling retries, timeouts, and circuit breaker patterns.
• Managing fallbacks, dead-ends, and designing recovery nodes.

Tool Usage and External Integrations

• Executing function and tool calls from within nodes and agents.
• Interacting with REST APIs and databases from the graph structure.
• 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 built-in safeguards.

Evaluation, Debugging, and Observability

• Tracing execution paths and inspecting node interactions.
• Using golden sets, evaluations, and regression tests for quality assurance.
• Monitoring quality, safety, cost, and latency metrics.

Packaging and Deployment

• Serving applications via FastAPI and managing dependencies.
• Versioning graphs and establishing rollback strategies.
• Developing operational playbooks and incident response procedures.

Wrap-up and Future Directions