AI Engineer Bootcamp
24 weeks — 12 weeks of coursework plus a 12-week industry project. From your first API call to a system in production, you build one delivery chain that can be observed, evaluated and iterated on.
Highlights
12 weeks of structured teaching plus a 12-week real company project you can put on a résumé
Every module ends in something that runs AND something that measures it — not a demo that worked once
Taught by working North American AI engineers; résumé review, mock interviews, referral network
Curriculum
1 · LLM engineering foundations: transformer intuition, tokenizers and context budgets, structured output, function calling
2 · Prompting and evaluation: system prompts, few-shot and chain-of-thought, turning a hunch into a measured score
3 · Retrieval-augmented generation: chunking, embeddings, choosing a vector store, hybrid search and reranking, citations
4 · Agents and the tool loop: ReAct, tool calling, task decomposition, retries and multi-agent coordination
5 · Memory and context engineering: session vs long-term memory, context compression, token cost control
6 · Model choice and fine-tuning: deciding whether to fine-tune at all, then LoRA and instruction tuning
7 · Shipping: inference serving, streaming responses, auth and rate limits, multi-tenancy, staged rollout
8 · Observability and cost: logs, tracing, evaluation dashboards, cost per request
9 · Safety and boundaries: prompt-injection defence, data leakage, output review
10–12 · Industry project: real requirements, architecture review, iterative delivery, Demo Day
Job sprint: rewriting your résumé around projects, AI system-design interviews, mock interviews and referrals
Core technologies
Languages & services
Models & SDKs
Agents & orchestration
Retrieval & storage
Evaluation & observability
Deployment