Leanpub Header

Skip to main content

AI Agents for Mathematicians

The instructor has published 100% of this course.Last updated on 2026-11-21

Learn to use AI agents for real mathematics, from computation and conjecture to Lean proofs and literature search, with one rule throughout: propose with the model, verify with a tool. A quiz in every lesson and a certificate.

Minimum price

$39.00

$59.00

You pay

Author earns

$

Also available for 1 course credit with a Learner Membership

PDF
EPUB
WEB
About

About

About the Course

AI Agents for Mathematicians is a practical online course for mathematicians, students, and educators who want to use AI agents without giving up rigour.

In 12 lessons you learn how language models, computer algebra systems, and proof assistants work together:

  • Foundations: what agents can and cannot do in mathematics, the three-part toolbox, and the gap between informal and formal mathematics.
  • Computation and discovery: computing rather than guessing, numerical exploration with symbolic confirmation, conjecture generation, and the counterexample hunt.
  • Proof: literature search grounded in arXiv and zbMATH, proof assistants and the trusted kernel, autoformalization and proof search in Lean.
  • Practice: a trust hierarchy for verification, research workflows, AI in teaching and learning, and where the field is heading.

The method throughout is simple: propose with the model, verify with a tool. Each lesson ends with a marked quiz (72 questions in total). Score 70% or more to earn your certificate of completion.

Share this course

Instructor

About the Instructor

Kompri

Kompri is an academic, author, investor, and educator based in Jambi, Indonesia. He holds a doctorate in Education Management and a bachelor’s degree in Physics, and has written and published dozens of books along with numerous other scholarly works. An avid reader, he is one of today’s most productive researchers. Alongside his academic career, he publishes both theoretical and practical books for readers around the world. His work rests on one conviction: knowledge is most valuable when it is clear, usable, and shared, so that anyone, in any country, can put it to work.

Material

Course Material

  • Welcome to the Course

  • How the Course Works

  • Earning the Certificate

  • Chapter 1: What AI Agents Mean for Mathematics

  • 1.1 The New Collaborator

  • 1.2 What Is an AI Agent?

  • 1.3 Why Mathematics Is Different: The Verification Advantage

  • 1.4 What Agents Can and Cannot Do for Mathematicians

  • 1.5 The Shape of This Book

  • In Practice — Silver at the Olympiad: AlphaProof and AlphaGeometry 2

  • Key Takeaways

  • References

  • Lesson 1 Quiz

    10 attempts allowed

  • Chapter 2: The Agent’s Toolbox: LLMs, Computer Algebra, and Proof Assistants

  • 2.1 Three Tools, Three Strengths

  • 2.2 Large Language Models: The Reasoning Engine

  • How you ask matters

  • 2.3 Computer Algebra Systems: Exact Computation

  • 2.4 Proof Assistants: Mechanical Certainty

  • In Practice — mathlib: a cumulative, verified library of mathematics

  • 2.5 Wiring the Tools Together: Tool-Calling

  • Key Takeaways

  • References

  • Lesson 2 Quiz

    10 attempts allowed

  • Chapter 3: Talking Mathematics to a Machine

  • 3.1 The Translation Problem

  • 3.2 How Formal Mathematics Looks

  • Reading a tactic proof

  • 3.3 Formalizing a Statement by Hand

  • 3.4 Autoformalization: Letting the Agent Translate

  • In Practice — The Liquid Tensor Experiment

  • 3.5 The Other Direction: Making Formal Mathematics Readable

  • Key Takeaways

  • References

  • Lesson 3 Quiz

    10 attempts allowed

  • Chapter 4: Agents for Symbolic and Numerical Computation

  • 4.1 Computation as the Agent’s Natural Habitat

  • 4.2 Driving a Computer Algebra System

  • 4.3 Numerical Exploration and Recognizing Constants

  • 4.4 Verifying Computational Results

  • 4.5 A Worked Example: From Problem to Verified Answer

  • In Practice — The BBP formula and experimental mathematics

  • Key Takeaways

  • References

  • Lesson 4 Quiz

    10 attempts allowed

  • Chapter 5: Conjecture and Discovery

  • 5.1 From Computation to Conjecture

  • 5.2 Generating Conjectures at Scale

  • 5.3 Testing Conjectures and Hunting Counterexamples

  • 5.4 The Division of Labour in Discovery

  • 5.5 A Worked Example: From Data to Conjecture

  • In Practice — The Ramanujan Machine

  • Key Takeaways

  • References

  • Lesson 5 Quiz

    10 attempts allowed

  • Chapter 6: Searching the Mathematical Literature

  • 6.1 Why Mathematical Search Is Hard

  • 6.2 How Agents Search: Retrieval and Grounding

  • 6.3 Reading and Synthesizing the Literature

  • 6.4 The Citation Trap: Hallucinated References

  • 6.5 A Worked Example: A Grounded Literature Review

  • In Practice — zbMATH Open

  • Key Takeaways

  • References

  • Lesson 6 Quiz

    10 attempts allowed

  • Chapter 7: Formal Proof Assistants and Why They Matter

  • 7.1 What a Proof Assistant Is

  • 7.2 The Trusted Kernel and the de Bruijn Criterion

  • 7.3 What Formalization Buys, and What It Costs

  • 7.4 mathlib and the Growth of Formalized Mathematics

  • 7.5 Why Proof Assistants Matter for Agents

  • In Practice — Flyspeck: formalizing the Kepler conjecture

  • Key Takeaways

  • References

  • Lesson 7 Quiz

    10 attempts allowed

  • Chapter 8: Agents that Prove: Autoformalization and Proof Search

  • 8.1 From Checking Proofs to Finding Them

  • 8.2 Autoformalization: From Informal to Formal

  • 8.3 Proof Search: How Agents Find Proofs

  • 8.4 Premise Selection: Finding the Right Lemmas

  • 8.5 The Realistic Workflow and Its Limits

  • In Practice — AlphaProof at the International Mathematical Olympiad

  • Key Takeaways

  • References

  • Lesson 8 Quiz

    10 attempts allowed

  • Chapter 9: Verifying and Trusting AI-Generated Mathematics

  • 9.1 The Trust Problem for AI Mathematics

  • 9.2 The Spectrum of Trust

  • 9.3 Two Questions: Is the Proof Correct, and Is the Statement Right?

  • 9.4 Verifying Informal AI Mathematics

  • 9.5 Responsible Use: Provenance, Disclosure, and Judgement

  • In Practice — FrontierMath and the honest measurement of AI mathematics

  • Key Takeaways

  • References

  • Lesson 9 Quiz

    10 attempts allowed

  • Chapter 10: Building a Mathematical Research Workflow with Agents

  • 10.1 From Capabilities to Workflow

  • 10.2 The Stages of Research and Where Agents Fit

  • 10.3 Orchestrating Agents and Tools

  • 10.4 Collaborative Formalization at Scale

  • 10.5 A Worked Research Session

  • In Practice — Formalizing the Polynomial Freiman–Ruzsa conjecture

  • Key Takeaways

  • References

  • Lesson 10 Quiz

    10 attempts allowed

  • Chapter 11: Teaching and Learning Mathematics with Agents

  • 11.1 Two Roles: Learner and Teacher

  • 11.2 The Agent as Tutor

  • 11.3 The Peril of Over-Reliance

  • 11.4 Teaching with Agents

  • 11.5 What Must Remain Human in Mathematical Education

  • In Practice — Khanmigo and the two-sigma aspiration

  • Key Takeaways

  • References

  • Lesson 11 Quiz

    10 attempts allowed

  • Chapter 12: The Road Ahead

  • 12.1 How Far We Have Come

  • 12.2 The Near-Term Trajectory

  • 12.3 The Frontier: Agents that Discover

  • 12.4 What Will Not Change

  • 12.5 Advice for the Mathematician

  • In Practice — FunSearch: when an agent discovered new mathematics

  • Key Takeaways

  • References

  • Lesson 12 Quiz

    10 attempts allowed

  • Glossary

  • Bibliography

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.)

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