Skip to content
Square1 Ai
On demandRecorded by an instructor, graded by Nova

Retrieval-Augmented Generation

Chunking, embeddings, reranking, citations, measured.

Retrieval-Augmented Generation is a self-paced online course: 9 hours of recorded sessions across 6 modules, with 5 graded by Nova, Square 1's AI tutor. The founding price is A$45, down from A$90. Start any time and keep your own pace.

Taught with
PythonClaudeSupabasePostgreSQL
Not sure it is for you?Take the free Generative AI skill check
Module by module
  1. Module 1Retrieval fundamentals60 min
  2. Module 2Ingestion and chunking90 min · graded
  3. Module 3Embeddings and hybrid search90 min · graded
  4. Module 4Reranking and generation with citations90 min · graded
  5. Module 5Evaluating RAG90 min · graded
  6. Module 6Project: a RAG application120 min · graded
recorded hours
9
modules, in order
6
deployed projects
4
graded checkpoints
5

Who it is for, and what you receive.

Developers in Python who have made one LLM API call and want a retrieval system they can measure.

Before you start

Python; one LLM API call before.

Skills
RAGembeddingsvector searchrerankingevals
  1. 1

    The recorded sessions

    Taught by an instructor, taken in order at your own pace, yours for twelve months.

  2. 2

    Graded work after every module

    Exercises and projects marked by Nova against a rubric you can read, with what you did well and what to fix.

  3. 3

    The projects

    Deployed and graded, each one on your record with its repository.

  4. 4

    Your hiring plan

    The same six-part plan the bootcamps use, run with the career agent.

  5. 5

    The record and the certificate

    Every grade at square1ai.com/u/{handle}; a credential ID that resolves at square1ai.com/verify.

  6. 6

    Nova as your tutor

    Help that is about your actual work, because it has read all of it.

The content plan, module by module.

9 recorded hours from an instructor, taken in order at your own pace. Nova grades the work at the end of each module.

  1. Lexical and semantic search

    Why RAG fails

    Measuring retrieval

The projects.

Each is deployed and graded by Nova against a rubric you can read before you start.

  1. Project 1

    An ingestion pipeline

    Take a public corpus, such as a state government's published planning guidelines, a mix of PDFs and web pages. Parse, chunk and tag each piece with source, section and date, so a rerun adds nothing twice.

    You hand in
    • Ingestion pipeline code
    • Chunks with metadata
    • Chunking strategy notes
    The rubric requires

    The pipeline reruns idempotently on the same corpus.

  2. Project 2

    A hybrid index

    Index your chunked corpus with both lexical and semantic search and combine them. Write labelled queries a planner would actually ask, including exact clause numbers, and measure recall at ten for each method and the hybrid.

    You hand in
    • Hybrid index
    • Labelled queries
    • Recall report by method
    The rubric requires

    Recall at ten is reported for lexical, semantic and hybrid retrieval.

  3. Project 3

    A cited answer engine

    Add a reranker and grounded generation so the engine answers planning questions with citations a reader can check against the source. It must decline when the corpus has no answer, and you measure faithfulness.

    You hand in
    • Reranking stage
    • Answer engine
    • Faithfulness eval
    The rubric requires

    Faithfulness is measured on a labelled set of questions.

  4. Project 4

    A deployed RAG application

    Ship the planning answer engine as a web app a member of the public could use, with its retrieval and faithfulness evals running against the deployed version and a failure analysis of the worst answers.

    You hand in
    • Deployed app URL
    • Eval suite
    • Failure analysis write-up
    The rubric requires

    The app is deployed with its eval suite passing.

What you can do at the end.

A RAG system you can measure and four graded projects on your record.

  1. 1

    Explain retrieval fundamentals and where RAG fails.

  2. 2

    Build ingestion and chunking pipelines.

  3. 3

    Build embeddings and hybrid search with measured recall.

  4. 4

    Rerank and generate cited answers with measured faithfulness.

  5. 5

    Evaluate a RAG system end to end and deploy it.

Roles this prepares you for
AI engineerSearch engineer

No placement rate is shown, because there are no graduates to count yet. The roles above are what the projects are built for.

Your record at the end.

Every exercise and project is graded by Nova against a rubric you can read, and every grade is kept on one page an employer can open and run. This is what the programme writes to it.

Graded, line by line
Nova reads every submission against the brief and the rubric and returns a score, what you did well and what to fix.
Module by module
Each module ends in graded work; the next opens when you are ready, on your own schedule.
Nova remembers
Help is about your actual work, because the tutor has every submission and every failed exercise of yours.
One page an employer can run
Every grade and project at /verify. An employer opens it and runs the code.
Record, Retrieval-Augmented GenerationExample
  1. Project

    An ingestion pipeline

    The pipeline reruns idempotently on the same corpus.

    Graded
  2. Project

    A hybrid index

    Recall at ten is reported for lexical, semantic and hybrid retrieval.

    Graded
  3. Project

    A cited answer engine

    Faithfulness is measured on a labelled set of questions.

    Graded
  4. Project

    A deployed RAG application

    The app is deployed with its eval suite passing.

    Graded

The entries, not the grades: those are yours to earn. The page lives at /verify and an employer needs no account to open it.

6 entries, at your own pace. Reserving a spot holds the founding price.

Pre-order

How we help you find a job.

Proof, not a certificate: the projects you deployed are the thing you show, and the tools below are yours to use.

The course booklet

The whole plan on a few pages: the parts, the projects, the gates and the price. We email it to you.

Get the booklet
  1. 1

    Your hiring plan

    The same six-part plan the bootcamps use: target roles, the gap map, the proof to send, weekly actions, an interview log, the outcome. You run it with the career agent.

  2. 2

    A record an employer can run

    Your graded projects on /verify. An employer opens it and runs the code.

  3. 3

    The career agent

    Paste a real job posting at /career and it maps the role to your graded work and what to do next.

  4. 4

    The roles directory

    Every role we prepare people for, what it pays and what it asks, at /roles.

  5. 5

    A path to the live cohort

    If you want the instructor, the gates and the hiring sprint, the bootcamp on the same subject is one waitlist away.

Questions people ask.

About Retrieval-Augmented Generation, answered from the plan on this page.

Further reading
Related programmes

Founding spots are open. Hold yours.

Recorded by an instructor who does this work, graded by Nova. Spot holders hear the opening date first, and nothing is charged before you confirm.

  • A RAG system you can measure
  • Four graded projects on your record
A$45A$90
Pre-order

About a minute. No account, no card, and nothing is charged until you confirm.

Pre-order · A$45Skill check