""Having a computer and being able to work it doesn't mean you can effectively apply computation or think computationally"
- Conrad Wolfram
Learn complex AI mathematics through simple pen‑and‑paper methods. No intimidating equations, just clear understanding with weekly live workshops.
""Having a computer and being able to work it doesn't mean you can effectively apply computation or think computationally"
- Conrad Wolfram
You're not alone. Thousands of Indian students and professionals feel overwhelmed by intimidating equations and abstract concepts.
Matrix multiplications, backpropagation, and gradient descent seem impossible to understand.
Premium courses cost ₹50,000+ with no guarantee of understanding the fundamentals.
Most resources are coding‑based and theoretical—there’s no step‑by‑step workbook to help learners understand the math behind AI.
Prof. Tom Yeh's proven methodology transforms complex AI mathematics into simple, hand‑drawable concepts that anyone can master.
“3 years ago, I made a controversial decision in my AI course: I ditched all Jupyter notebooks for whiteboard writing and drawing by hand… They brought their own paper notebook, followed my whiteboard drawing step by step by hand. That's how they learned in India—paper first, computers later.”
— Prof. Tom Yeh
Sania Rahaman shared her sketch of MCP vs Function Calling at a recent livestream event.
“I remember one day watching professor Tom Yeh lecture on YouTube and thinking, I wish I could meet him one day… Had a misunderstanding that function calling was bad in the beginning but… they’re both majestic in their own way depending on the situation.”
Ravi Vats shared his note about NdLinear at the AI by Hand SOTA workshop this week.
“I recently was part of a SOTA paper study session… titled ‘NdLinear Is All You Need for Representation Learning’… When we flatten a matrix of pixels… the spatial relationship between them fails to be mathematically represented.”
Gurjas Kapur completed 500+ exercises in the Deep Learning Math Workbook India Edition.
“Last week, I received a signed copy of the Deep Learning Math Workbook – AI by Hand (India Edition)… I have completed 500+ exercises and finished the entire workbook! … It covers 12 hands‑on chapters with around 50 exercises each, including: Dot Product, Matrix Multiplication…”
Experience the most effective way to learn deep learning mathematics with our unique approach.
Interactive workshops every week with real-time Q&A and personalized guidance in IST timezone.
Examples and case studies relevant to Indian students with cultural context and familiar scenarios.
Complex concepts broken down into digestible, visual steps that build understanding progressively.
World-class instruction combined with service and kindness for skilling people worldwide.
Professor at University of Colorado Boulder with 15+ years of experience making complex AI concepts accessible.
Collaborating with Prof.Tom Yeh and currently working as a teacher in weekly workshops on AI by hand.
Join our live interactive sessions every week. All times in IST for your convenience.
Draw your first neural network by hand
Calculate self‑attention step by step
Understand the latest AI trends
Real-time Q&A with Prof. Tom Yeh and his team
Step‑by‑step math exercises to do with pen and paper
Get your hand‑drawn solutions reviewed during class
Learn side‑by‑side in Gather.town with avatars at five rooftop tables—no Zoom isolation
See how modern AI hardware performs matrix operations through our interactive hand-drawn visualization.
Follow Prof. Tom Yeh for the latest updates: Substack Magazine | LinkedIn Profile
The Deep Learning Math Workbook – India Edition by Prof. Tom Yeh and Dr. Anish Roy Chowdhury is now available and tailored for learners in India. This edition emphasizes affordability and accessibility, aligning with traditional Indian learning methods that prioritize foundational understanding through handwritten exercises.
This workbook is a streamlined version of the original 312‑page edition, focusing on core mathematical concepts essential for deep learning. It includes exercises on dot products, matrix multiplication, linear layers, softmax functions, gradients, and more. Designed for self‑paced learning with pencil and paper, it's ideal even in environments without advanced computing resources.
Start your AI journey today with our comprehensive free resources.
🌐 Official Website | 📰 Imagine AI Substack (3K+ subs) | ▶️ YouTube Channel
📩 For Educators, Institutions, NGOs: Connect with us to bring AI by Hand to your organization.
🤝 Volunteer / Ambassador Program: Open for students and teachers. Apply here.