Skip to main content

Who this track is for

The AI Engineer track is built for developers who want to ship AI products. You’ll learn how to work with large language models, build retrieval-augmented generation systems, design multi-agent architectures, and deploy AI systems that actually work in production. Best fit if you:
  • Come from a software engineering background
  • Want to work at AI-native startups or build your own AI product
  • Are interested in the systems side of AI, not just research

Curriculum

Level 1 — LLM Foundations (free)

ModuleKey Topics
Working with LLMsAPIs, tokens, context windows, system prompts, few-shot prompting
Prompt EngineeringChain-of-thought, structured output, evaluation, red-teaming

Level 2 — Building with LLMs

ModuleKey Topics
RAG FundamentalsChunking, embeddings, vector search, pgvector, retrieval quality
LangChain & PipelinesChains, memory, document loaders, evaluation
Structured Output & Function CallingJSON mode, tool use, schema validation

Level 3 — Production AI Systems

ModuleKey Topics
AI AgentsReAct loop, tool use, agent memory, multi-step reasoning
Multi-Agent SystemsOrchestration, specialised agents, handoff protocols
LLM EvaluationBenchmarking, LLM-as-judge, regression testing

Level 4 — Advanced AI Engineering

ModuleKey Topics
Fine-tuning & AlignmentLoRA, RLHF basics, dataset curation, evaluation
AI InfrastructureCost management, caching, streaming, observability
Production PatternsRate limiting, fallbacks, human-in-the-loop, guardrails
Multimodal SystemsVision, audio, document understanding

Level 5 — Capstone

Build a complete, production-ready AI product — from architecture design through deployment. Peer-reviewed by your cohort.

What you’ll be able to build

  • RAG systems — document Q&A, semantic search, knowledge bases with real retrieval metrics
  • AI agents — autonomous systems that use tools, maintain memory, and complete multi-step tasks
  • LLM pipelines — evaluation frameworks, structured output, production-grade APIs
  • Complete AI products — end-to-end systems with proper error handling, observability, and cost controls