Course Objective & Scope
Getting a prototype agent running takes a few lines of code; the hard part is building one that stays stable on long tasks, keeps manageable state and has traceable failures. The programme compares graph-based agent architecture with LangGraph against rapid development with the OpenAI Agents SDK and the Claude Agent SDK.
For a task adapted to their organisation (for example research and report drafting), participants build an agent that uses tools, persists state via checkpoints and enforces step and budget limits, then add tracing with LangSmith or Langfuse and package it as a container for deployment.
Who Should Attend?
- AI engineers and LLM application developers
- Backend and platform teams building agent platforms
- Technical leads and solution architects
- Prerequisite: solid Python, LLM API and tool calling experience
Key Takeaways
- Build an agent with LangGraph from state, nodes and conditional edges
- Choose a framework based on the strengths and weaknesses of agent SDKs
- Achieve controlled autonomy with step, budget and time limits
- Monitor and debug agent decisions with tracing
- Package the agent as an API and prepare it for deployment
Curriculum
Agent Architectures and Framework Selection
- ReAct, plan-and-execute and graph-based approaches
- Comparing LangGraph, the OpenAI Agents SDK and the Claude Agent SDK
Stateful Agents with LangGraph
- State, nodes, edges and conditional routing
- Persistent state and resumption with checkpointers
Tools and External Systems
- Tool registration and connecting to MCP servers
- Sandboxes for code execution and file operations
Controlled Autonomy
- Step, budget and time limits; stop conditions
- Preventing loops and repeated failures
Tracing and Debugging
- Run traces with LangSmith and Langfuse
- Analysing failed runs
Preparing for Deployment
- Serving the agent as an API and packaging it in a container
- Async tasks, queues and secrets management
Other Courses in This Category
-
Technical programme on how agents plan and execute multi-step tasks: ReAct, plan-and-execute, task decomposition, reflection and replanning, using reasoning models, and loop termination strategies.
- Live Course
- 6 modules
-
Agent memory and context management for long-horizon tasks and multi-session interactions: working memory, episodic and semantic long-term memory, context compaction, and persistent memory architecture with Mem0 and LangGraph Store.
- Live Course
- 5 modules
-
Multi-agent architectures that split complex work across specialised agents: supervisor, hierarchical and swarm patterns, comparing LangGraph, CrewAI and AutoGen, the A2A protocol and shared state management.
- Live Course
- 5 modules
-
Human-in-the-loop agent architecture that preserves human oversight on critical actions: interrupt/resume flows, risk-based approval thresholds, escalation queues and agents that improve from human feedback.
- Live Course
- 5 modules
-
Connecting enterprise databases, APIs and internal systems to LLMs and agents cleanly and securely with Model Context Protocol (MCP): MCP server/client development, authorisation and enterprise integration architecture.
- Live Course
- 5 modules
-
Turning institutional knowledge, procedures and tools into modular, reusable agent skills: SKILL.md structure, progressive disclosure, templates, scripts and enterprise skill library governance.
- Live Course
- 5 modules
Get in touch
