Terminus Expanse
CurriculumBlogPricingSign in
Back to catalog
Artificial intelligence

LLMs, Embeddings & RAG

What large language models actually do, and how to ground them in real data.

Prerequisite: Neural Networks

What you'll actually learn

Large language models get talked about like magic right up until you need to actually build something reliable with one — then embeddings, retrieval, and grounding stop being buzzwords and become the specific tools that decide whether your system answers from real data or confidently makes something up. This track builds from what a model is actually doing when it predicts the next token, up through how embeddings turn text into something you can search by meaning, to a real retrieval-augmented generation pipeline.

What you'll be able to do

You'll build a RAG pipeline that answers questions grounded in a real document set instead of the model's untethered guesses, and be able to explain exactly why it sometimes still gets something wrong — which is the actual skill gap between "I called an LLM API" and being trusted to build something people rely on for a real answer.

Syllabus

1Tokens: Why an LLM Does Not See Wordscore~9 min
2Next-Token Prediction: The One Thing an LLM Actually Doescore~9 min
3Attention: How a Model Decides What Matterscore~10 min
4Embeddings: Meaning as a Position in Spacecore~10 min
5Semantic Search: Finding Meaning, Not Just Keywordscore~9 min
6RAG: Grounding an LLM in Real Documentscore~10 min
7Prompt Engineering: A Real Skill, Not a Trickcore~9 min
8The Context Window: A Model’s Limited Working Memorycore~8 min
9Capstone: Design a RAG-Powered Support Assistant, On Papercore~12 min

Leads to

Serving AI in Production

Frequently asked

How long does LLMs, Embeddings & RAG take to complete?

Roughly 1 hour across 9 hands-on quests — you can go at your own pace and pick up exactly where you left off.

What do I need to know before starting LLMs, Embeddings & RAG?

You should be comfortable with Neural Networks first — the skill tree unlocks LLMs, Embeddings & RAG once you've cleared those.

Is LLMs, Embeddings & RAG free to learn?

Yes — LLMs, Embeddings & RAG is fully available on the free tier, starting with a free first quest and no payment method required to sign up. Plus and Elite remove pacing limits but don't gate any of the LLMs, Embeddings & RAG curriculum behind a paywall.

Start this skill for free

Create a free account and begin your first quest — no card required.

Start free