RAG Engineer roadmap › Courses
Best courses for RAG Engineers
This list covers the courses that take you from a basic retrieval pipeline to a measured one. It starts with the stages every RAG system has, moves to embedding and reranker vocabulary, and ends with advanced retrieval methods and their evaluation.
Most of the material is free. The DeepLearning.AI course is free for a limited time during its platform beta, as its page states, so check the current terms before you plan around it.
Build a plain pipeline first, then use the advanced course to change one retrieval method at a time. The value is in the evaluation loop, not in any single technique.
- 5 items in 3 stages: 5 courses and tutorials, 0 certifications.
- We opened every page on November 20, 2026 and show only what the provider states.
- No prices, ratings or enrollment counts.
See the full RAG Engineer roadmap →Best books for RAG Engineers →
How to choose your courses
A short method, so you spend your hours on the right thing for where you are instead of collecting certificates.
Find your stage
Getting started · 2 items
The stages of RAG and the first end-to-end pipeline. Read these before you change any part of retrieval, so you can name the stage that failed.
Jump to this stage ↓Doing the job · 2 items
The model and evaluation layer: the embedding and transformer vocabulary, and the advanced retrieval course with its evaluation metrics.
Jump to this stage ↓Senior and leading · 1 items
Two-stage retrieval. A bi-encoder finds candidates and a cross-encoder reranks them, which is the shape production systems often take.
Jump to this stage ↓
If you are unsure, start one stage lower than you think. Later items assume what the earlier ones teach.
Begin with one course
Our starting pick
Introduction to RAG
LlamaIndex
It maps the loading, indexing and querying stages so every later course has a place to attach. It is free and short.
See the course ↓Then pick by your situation
Choose the next course by the problem in front of you, not by its place in the list.
If your first retrieval results look plausible but you cannot say whether they got better
take Building and Evaluating Advanced RAG (doing the job)
If you need the transformer vocabulary before you tune an embedding model
take LLM Course (doing the job)
If your first-stage search finds candidates but the top answers are the wrong ones
take Retrieve & Re-Rank (senior and leading)
Free path or paid path?
5 of the 5 courses and tutorials can be read or taken without paying, so you can build most of the skill at no cost. Check each certification page for exam fees.
- Building and Evaluating Advanced RAG
- Retrieval Augmented Generation
- Introduction to RAG
- LLM Course
- Retrieve & Re-Rank
Paid courses earn their place when you need structure, deadlines or feedback. Check what the provider says is included before you commit.
Set a time budget
| Stage | Items with a stated length | Hours of material | At 4 hrs a week |
|---|---|---|---|
| Getting started | 0 | none stated | n/a |
| Doing the job | 1 | 2 hrs | 1 wks |
| Senior and leading | 0 | none stated | n/a |
| All stages | 1 | 2 hrs | 1 wks |
Totals are approximate.
Exam time is not included, and hands-on practice takes longer than the material itself. Treat this as a floor.
Judge any course before you commit
Who runs it
Prefer courses from the people who build or govern the tool. They tend to stay current, and an exam from a tool’s own maker is the clearest credential for that tool.
Hands-on or watch-only
In this field you learn by running things. Choose material where you build a pipeline, a cluster or a dashboard over material you only watch.
How current it is
Tools change fast. Look for a version or last-updated date on the provider’s page, and check anything several years old against the current docs.
What you will be able to show
Ask what you will have at the end: a working repository, a deployed service or a passed exam. If it is only a certificate of completion, pair it with a project of your own.
Getting started
The stages of RAG and the first end-to-end pipeline. Read these before you change any part of retrieval, so you can name the stage that failed.
What you will learn
- The loading stage
- The indexing stage
- The querying stage
- Important concepts within RAG
Good for
A clear map of the stages every RAG system has, so you can name the stage that failed when an answer is wrong.
Also stated by the provider
- The docs.llamaindex.ai URL redirects to this page
- Hands-on tutorialFree to learn
Retrieval Augmented Generation (RAG) with Deep Agents
LangChain
View course →What you will learn
- Index documentation into a vector store
- Build a search tool over the index
- Delegate analysis to subagents
- Synthesize the final answer
Good for
An end-to-end RAG pipeline with agents. Read it after you have built a plain pipeline of your own.
Also stated by the provider
- The python.langchain.com tutorial URL redirects to this page
Doing the job
The model and evaluation layer: the embedding and transformer vocabulary, and the advanced retrieval course with its evaluation metrics.
- CourseFree to learn
Building and Evaluating Advanced RAG
DeepLearning.AI
- 2 hours of material
What you will learn
- The advanced RAG pipeline
- Sentence-window retrieval
- Auto-merging retrieval
- The RAG triad of evaluation metrics
Good for
Retrieval methods beyond one vector search, with the evaluation loop that shows whether each change helps.
Also stated by the provider
- Free for a limited time during the DeepLearning.AI platform beta, per the page
- Page lists a 2 hour 5 minute course duration
- CourseFree to learn
LLM Course
Hugging Face
- About 6 to 8 hours of work per week, one chapter per week
What you will learn
- How the pipeline function runs NLP tasks
- The Transformer architecture
- Encoder, decoder and encoder-decoder models
Good for
The model vocabulary behind embeddings and rerankers. Work through the embedding chapters before you tune one.
Also stated by the provider
- Open course on language models, published on Hugging Face
Senior and leading
Two-stage retrieval. A bi-encoder finds candidates and a cross-encoder reranks them, which is the shape production systems often take.
What you will learn
- Dense retrieval with a bi-encoder
- Re-ranking with a cross-encoder
- Lexical search as a baseline
Good for
A two-stage retrieval pattern you can run locally. Many production RAG systems end up using this shape.
Also stated by the provider
- Runs locally with the sentence-transformers library
- Public project documentation
Details such as hours, format and whether material is free are as the providers state them on November 20, 2026, and can change. Confirm on the provider’s page before you commit. See our disclosure.
Frequently asked questions
Which course should a RAG Engineer take first?
Introduction to RAG, from LlamaIndex. It maps the loading, indexing and querying stages so every later course has a place to attach. It is free and short.
Can a RAG Engineer learn this for free?
Largely, yes. 5 of the 5 courses and tutorials here can be read or taken without paying (Building and Evaluating Advanced RAG, Retrieval Augmented Generation, Introduction to RAG, LLM Course and more). Check each certification page for exam fees: not every page states one.
How long does it take?
The providers state 2 hours of material across 1 of the 5 courses and tutorials; the rest do not state a length. At four hours a week that is about 1 weeks for the stated material alone, before projects and exam preparation.
How was this list put together?
We opened each course, tutorial and certification page on November 20, 2026 and wrote down only what the page states: who runs it, what it covers, hours and exam format where given, and whether the material is free to read. We do not show prices, ratings or enrollment figures. Details can change, so confirm on the provider’s page before you commit.