← All programs
4 weeks · Intermediate–Advanced
LLM Engineering Intensive
For engineers who already code — a focused 4-week sprint on building, evaluating, and shipping production-grade LLM systems.
What you'll learn
Production LLM engineering
Advanced Prompting
Chain-of-thought, few-shot, system prompts, and structured output techniques.
RAG Architecture
Embeddings, vector databases, chunking strategies, and hybrid retrieval.
Fine-tuning
When and how to fine-tune — LoRA, QLoRA, and RLHF basics.
Evaluation
Build evals that tell you whether your system actually improved.
Agents & Tool Use
Build autonomous agents that use tools, search, and APIs.
Production Ops
Latency, cost, rate limits, caching, and observability in production.
Program structure
4 weeks, week by week
1
Week 1: LLM internals
Transformers, tokenization, context windows, and model selection.
2
Week 2: RAG systems
Build a production RAG pipeline with real retrieval, re-ranking, and evals.
3
Week 3: Agents
Multi-step agents, tool use, and memory patterns.
4
Week 4: Ship it
Deploy your LLM system, instrument it, and present to the cohort.