Skip to main content
Advanced8 weeks3 modules

LLM Engineering

Build production systems on top of language models

The skills companies are hiring for right now: prompting as engineering, retrieval-augmented generation, agents with tool use, and rigorous evaluation. You will build a complete RAG application and an agent system, and learn how to measure whether they actually work.

Curriculum

  1. 1Working with LLMs

    • How transformers work: an engineer's mental model
    • Prompting patterns: few-shot, chain-of-thought, structured output
    • Streaming, function calling, and API design
    • Cost, latency, and context-window budgeting
  2. 2Retrieval-Augmented Generation

    • Embeddings and semantic search
    • Chunking strategies that preserve meaning
    • Vector stores, hybrid search, and re-ranking
    • Grounding, citations, and hallucination control
  3. 3Agents & Evaluation

    • Agent loops: planning, tool use, and recovery
    • Guardrails, sandboxing, and safety boundaries
    • Evaluation suites: golden sets, LLM-as-judge, regression tests
    • Capstone: documentation assistant with RAG + tools