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 Agentic Design Patterns: An Applied Guide
- Comparing graph-based structures with linear sequences: Use cases and justification
- Orchestrating agents, tools, and planner-executor cycles
- Establishing a baseline: A minimal agentic graph implementation
State Management, Memory Architecture, and Context Transfer
- Architecting graph state and node interface definitions
- Distinguishing between ephemeral and persistent memory storage
- Managing context windows, summarization techniques, and state restoration
Decision Logic and Execution Control
- Implementing conditional routing and multi-path decision making
- Configuring retry mechanisms, timeout policies, and circuit breaker patterns
- Managing fallback paths, handling dead ends, and defining recovery states
Tool Integration and External System Interoperability
- Executing function and tool calls from nodes and agents
- Consuming RESTful APIs and database resources within the graph structure
- Processing, parsing, and validating structured data outputs
Retrieval-Augmented Generation (RAG) in Agentic Workflows
- Strategies for document ingestion and text chunking
- Utilizing embeddings and vector databases such as ChromaDB
- Ensuring response accuracy through citations and safety guardrails
Quality Assurance, Troubleshooting, and System Observability
- Tracing execution paths and analyzing node-level interactions
- Employing golden datasets, evaluation frameworks, and regression testing
- Monitoring performance metrics, safety compliance, and resource costs
Deployment, Packaging, and Lifecycle Management
- Serving applications via FastAPI and managing dependencies
- Versioning graph definitions and implementing rollback procedures
- Developing operational playbooks and incident response protocols
Conclusions and Recommended Progression
Requirements
- Proficiency in Python programming
- Experience developing LLM applications or managing prompt chains
- Working knowledge of RESTful APIs and JSON data structures
Target Audience
- AI engineers
- Product managers
- Developers responsible for building interactive LLM-driven systems
