A sixteen-week, engineering-first path from LLM fundamentals to production multi-agent systems — RAG, LangChain, LangGraph, MCP and A2A protocols, evaluation, safety and deployment.
AI systems in Python, LLM APIs, how LLMs behave.
Function calling, structured outputs, cost/latency basics.
Embeddings, vector databases, hybrid retrieval, reranking.
LangChain, LangGraph, tool use, agent memory.
MCP, A2A, orchestrator-worker patterns.
Eval suites, observability, guardrails, prompt-injection defence.
Docker, sandboxed execution, production monitoring.
Documentation, demos, stakeholder presentation.
One agentic system, start to finish — deployed and monitored.
Senior Data Scientist & AI/ML Programme Leader, 11 years of experience. Instructor at Berkeley Haas and Imperial College Business School. Every batch — up to 50 learners — taught live, session by session.
Every learner finishes with a deployed, portfolio-ready project — not a certificate for finishing videos.
Full curriculum, week-by-week session plan, capstone tracks and mentor details — sent straight to your inbox and available to download immediately.
Message us on WhatsApp or get in touch — we'll help you confirm this is the right programme before you commit to anything.