Kick off your book project, get started with GhostAI, get better at marketing, or spend the day doing all three! Free live workshops on Zoom. Saturday, June 27, 2026.

Leanpub Header

Skip to main content

Vectors & Embeddings: The One Idea Every AI System Is Built On

Learn vectors & embeddings from zero and build a real AI semantic search engine. Free tools, no math fear. The first skill of every AI engineer.

The instructor has published 100% of this course.Last updated on 2026-06-26

ChatGPT's memory. AI search. RAG. Recommendation engines. All of it stands on ONE idea — and almost nobody using AI actually understands it. This course takes you from absolute zero (no code, no math, no fees) to building your own AI search engine that finds things by meaning. In the age of AI, this is the line between people who prompt and people who build. Which side are you on?

Minimum price

$45.00

$59.00

You pay

Author earns

$
PDF
EPUB
WEB
About

About

About the Course

While everyone else is typing prompts, you'll understand what's beneath the surface.


Every AI product you touched this week- the chatbot that remembered your documents, the search box that understood what you meant, the app that recommended exactly the right thing — runs on a single idea: turning meaning into numbers. Vectors and embeddings.

Here's the uncomfortable truth of the AI age: there are now two kinds of people. The millions who use AI through a text box — and the small group who understand the machinery underneath and get paid like it. Every AI engineer, every RAG developer, every AI-native founder crossed the same first bridge, and it's this one. Embeddings are where real AI engineering starts. Skip this, and every tutorial, every job posting, every architecture diagram stays slightly out of reach — forever explained in words you nod along to but couldn't build.

This course takes you across that bridge from absolute zero. No programming background. No math beyond school arithmetic — and we re-teach even that from scratch. No paid accounts, no API keys, no GPU: everything runs free in your browser.

And you don't just read — you build. By the end you'll have made, with your own hands and understanding every single line:

- A semantic search engine that finds documents by meaning — your query shares zero words with the result, and it still finds it

- A smart FAQ bot that answers customers however they phrase the question

- The judgment most tutorials never teach: the gotchas that fool even professionals — like why "returns accepted" and "returns NOT accepted" look nearly identical to an AI, and what the pros do about it

Fifteen hand-held lessons. Two hundred quiz questions with instant feedback, thirty-six exercises, eleven ready-to-run notebooks. A certificate when you pass.

AI is not waiting. Every month you delay, the gap between the people who understand this and the people who don't gets wider — and more expensive. Cross the bridge now, while being early still counts.

Start the first lesson. It begins with a single arrow.

Instructor

About the Instructor

Ritesh Modi

Ritesh Modi is Head of AI at MarketOnce and a former Forward Deployed Engineer at Microsoft. He has spent more than a decade building and shipping production systems across cloud, distributed computing, and applied machine learning, working with organizations ranging from global enterprises to fast-moving startups. His recent work focuses on applied large language models, designing systems that turn pretrained models into reliable, task-specific tools.

Ritesh has authored multiple technology books and speaks regularly at industry conferences on AI, cloud architecture, and software engineering. His writing philosophy rests on a simple belief: the best technical books are written by practitioners who still remember what it felt like to not understand something, not by experts who have forgotten. Every explanation in this book was tested against that standard, if it would not have made sense to him when he was first learning this material, it was rewritten until it did.

He writes, shares ideas, and connects with readers at www.riteshmodi.com. When he is not writing or building AI systems, he can be found mentoring engineers, exploring new architectures, or debugging a training run that should have converged three hours ago.

Material

Course Material

  • Welcome: Numbers That Understand Meaning

  • What you’ll be able to do at the end

  • Every lesson builds a part of the final machine

  • How this course works

  • Exercise: How Your Score Is Counted

  • You’ll write a little code — here’s the deal

  • A promise

  • Read on…

  • Setup: Pick How You’ll Run the Code

  • First, three words you’ll see a lot

  • Choose ONE of three ways to run the code

  • Track A: Google Colab (recommended, nothing to install)

  • Track B: VS Code (a real editor on your computer)

  • Track C: Local Jupyter (the classic notebook)

  • Run the Smoke Test

  • Try one small change

  • Exercise: Does Every Word Become 384 Numbers?

  • Check your understanding

  • Quiz: Setup Check

    3 attempts allowed

  • Read on…

  • What Is a Vector?

  • The intuition first (no math yet)

  • The idea, made precise

  • The smallest possible example

  • Now in code

  • Your turn

  • Exercise: Make Your Own Vector

  • Exercise: Build the Vector I’m Thinking Of

  • Common mistakes (and how to fix them)

  • Check your understanding

  • Quiz: Vectors

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • Combining Arrows: Vector Arithmetic

  • The intuition: adding is just “one walk, then another”

  • Made precise

  • Subtraction: the arrow that takes you from A to B

  • Stretching: scalar multiplication

  • Now in code

  • Your turn

  • Exercise: Make the Sum Come Out Right

  • Exercise: The Arrow Between Two Points

  • Exercise: Fix the Broken Cell

  • Common mistakes (and how to fix them)

  • Check your understanding

  • Quiz: Vector Arithmetic

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • How Long Is an Arrow? Magnitude and Distance

  • The intuition: blocks walked vs. straight-line distance

  • Check it yourself on grid paper

  • Made precise: square, add, square root

  • Now in code

  • Distance between two points

  • Your turn

  • Exercise: Lengths — Forwards, Then Backwards

  • Exercise: Distance Between Two Points

  • Exercise: The Ruler, Forwards and Backwards

  • Common mistakes (and how to fix them)

  • Check your understanding

  • Quiz: Magnitude and Distance

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • Words as Arrows: Dot Product and Cosine Similarity

  • Where we are, and the big idea

  • The intuition: two people walking

  • Step 1: the dot product

  • Step 2: from dot product to cosine similarity

  • Working one cosine fully by hand

  • Why dividing by lengths cancels length

  • Now in code: the function we’ll reuse all course

  • Seeing a negative cosine

  • Words as arrows

  • A first taste of search: nearest neighbours

  • Your turn

  • Exercise: Dot Products by Hand

  • Exercise: One Cosine, All the Way by Hand

  • Exercise: Write Your Own Function

  • Exercise: Fix the Broken Ruler

  • Exercise: Add a Word, Find Its Neighbours

  • Common mistakes (and how to fix them)

  • Check your understanding

  • Quiz: Cosine Similarity

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • Seeing Higher Dimensions

  • The intuition: a spreadsheet row is a vector

  • Nothing breaks

  • The problem: how do you see it?

  • The trick: PCA (a photograph of the data)

  • Your turn

  • Exercise: Predict, Then Check

  • Exercise: Add Your Own Words to the Map

  • Common mistakes (and how to fix them)

  • Check your understanding

  • Quiz: Higher Dimensions

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • What Is an Embedding?

  • What we did, and the one thing we faked

  • What a neural network is (in plain words)

  • How does it invent its own directions?

  • The reveal

  • The map of meaning

  • Two flavours: word embeddings and sentence embeddings

  • A famous party trick

  • Your turn

  • Exercise: Say It In Your Own Words

  • Exercise: Word Embedding or Sentence Embedding?

  • Common mistakes (and how to fix them)

  • Check your understanding

  • Quiz: What Is an Embedding?

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • Real Embeddings, Free and Local

  • The model we’ll use

  • One line to embed text

  • Embedding many sentences at once

  • From many words to ONE vector

  • Know your model: the fact sheet

  • Searching by meaning

  • Seeing them on the map

  • Your turn

  • Exercise: Embed Your Own Sentences

  • Exercise: Try to Break the 384

  • Exercise: Watch Context Steer the Word “Bank”

  • Exercise: Find the Cut-Off

  • Exercise: Context Steers the Word “Spring”

  • Common mistakes (and how to fix them)

  • Check your understanding

  • Quiz: Real Embeddings

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • A Home for Your Vectors: A Local Vector Store

  • The intuition: a library with a magic catalogue

  • What the store does for you

  • Creating the store

  • Giving each item an id

  • Adding and querying

  • A small thing about the score

  • Reading the result: what is that [0]?

  • It remembers

  • Your turn

  • Exercise: Steer the Search

  • Exercise: Reopen a Saved Store

  • Exercise: Fix the Bracket Bug

  • Exercise: Turn a Distance Into a Similarity

  • Exercise: Add New Sentences to the Store

  • Common mistakes (and how to fix them)

  • Check your understanding

  • Quiz: Vector Store

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • Guided Capstone: Build Your Semantic Search Engine

  • Step 1: a small library of documents

  • Step 2: give each document an id

  • Step 3: create the store and add the documents

  • Step 4: the search function

  • Step 5: the magic — searching with no shared words

  • Your turn

  • Exercise: Search Your Own Library

  • Exercise: When Nothing Matches

  • Exercise: Predict the Top Two

  • Common mistakes (and how to fix them)

  • Check your understanding

  • Quiz: Semantic Search

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • Mini Capstone: Build One Yourself

  • What you’re building: a Smart FAQ Finder

  • The key idea: store the questions, attach the answers

  • Your checklist

  • Hints (open only if you’re stuck)

  • What success looks like

  • The full sample solution

  • Your turn (graded)

  • Exercise: Your FAQ Finder

  • Exercise: Add a New FAQ Without Rebuilding

  • Exercise: Produce Exactly These Ids

  • Exercise: When the FAQ Has No Answer

  • Common mistakes (and how to fix them)

  • Check your understanding

  • Quiz: Smart FAQ Finder

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • How the Model Reads: Tokens and the Four Dials

  • The intuition: Lego bricks of language

  • See it with the real tokenizer

  • Real text is weirder than words

  • The four dials

  • Your turn

  • Exercise: Take the Tokenizer for a Spin

  • Exercise: The Most Expensive Short Phrase

  • Exercise: The Storage Bill

  • Common mistakes (and how to fix them)

  • Check your understanding

  • Quiz: Tokens and the Four Dials

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • When Embeddings Fool You: Five Gotchas

  • Gotcha 1: there is no single “sentence embedding”

  • Gotcha 2: the silent cut-off

  • Gotcha 3: rare names shatter

  • Gotcha 4: a similarity score is not a grade

  • Gotcha 5: embeddings barely see “not” — or numbers

  • The five on one card

  • Your turn

  • Exercise: Catch the Negation Trap Yourself

  • Exercise: Two Days, Twenty Days

  • Exercise: Read the Label — Then Imagine It Different

  • Exercise: Why “Above 0.8” Is Not Portable

  • Common mistakes (and how to recognise them)

  • Check your understanding

  • Quiz: The Five Gotchas

    3 attempts allowed

  • Recap in plain words

  • Read on…

  • Final Quiz and Where Vectors Take You Next

  • How far you’ve come

  • The final quiz

  • Final Quiz

    3 attempts allowed

  • An honest word on the limits

  • Where this same idea goes next

  • Where to go from here

  • Your certificate

  • Congratulations

The Leanpub 60 Day 100% Happiness Guarantee

Within 60 days of purchase you can get a 100% refund on any Leanpub purchase, in two clicks.

See full terms...

Earn $8 on a $10 Purchase, and $16 on a $20 Purchase

We pay 80% royalties on purchases of $7.99 or more, and 80% royalties minus a 50 cent flat fee on purchases between $0.99 and $7.98. You earn $8 on a $10 sale, and $16 on a $20 sale. So, if we sell 5000 non-refunded copies of your book for $20, you'll earn $80,000.

(Yes, some authors have already earned much more than that on Leanpub.)

In fact, authors have earned over $15 million writing, publishing and selling on Leanpub.

Learn more about writing on Leanpub

Free Updates. DRM Free.

If you buy a Leanpub book, you get free updates for as long as the author updates the book! Many authors use Leanpub to publish their books in-progress, while they are writing them. All readers get free updates, regardless of when they bought the book or how much they paid (including free).

Most Leanpub books are available in PDF (for computers) and EPUB (for phones, tablets and Kindle). The formats that a book includes are shown at the top right corner of this page.

Finally, Leanpub books don't have any DRM copy-protection nonsense, so you can easily read them on any supported device.

Learn more about Leanpub's ebook formats and where to read them

Write and Publish on Leanpub

You can use Leanpub to easily write, publish and sell in-progress and completed ebooks and online courses!

Leanpub is a powerful platform for serious authors, combining a simple, elegant writing and publishing workflow with a store focused on selling in-progress ebooks.

Leanpub is a magical typewriter for authors: just write in plain text, and to publish your ebook, just click a button. (Or, if you are producing your ebook your own way, you can even upload your own PDF and/or EPUB files and then publish with one click!) It really is that easy.

Learn more about writing on Leanpub