# TinyTorch: Build ML Systems from Scratch

<div style="text-align: center; margin: 4rem 0;">

## Don't just import it. Build it.

**A Comprehensive Course in Machine Learning Systems Engineering**

---

**Prof. Vijay Janapa Reddi**
Harvard University

---

**2025 Edition**

</div>

---

## About This Book

This book provides a comprehensive, hands-on introduction to building machine learning systems from scratch. Rather than treating ML frameworks as black boxes, you'll implement every component yourself—from tensors and gradients to optimizers and attention mechanisms—gaining deep understanding of how modern ML systems actually work.

## Course Philosophy

**Build → Profile → Optimize**

You'll implement each system component, measure its performance characteristics, and understand the engineering trade-offs that shape production ML systems. This approach transforms you from a framework user into a systems engineer who can debug, optimize, and architect ML solutions at scale.

## Three-Tier Learning Pathway

**Foundation Tier (Modules 01-07)**: Build mathematical infrastructure
**Architecture Tier (Modules 08-13)**: Implement modern AI architectures
**Optimization Tier (Modules 14-20)**: Deploy production systems

## Course Website

For the latest updates, interactive exercises, and community resources, visit:

**https://www.mlsysbook.ai/tinytorch**

## License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

---

© 2025 Vijay Janapa Reddi. All rights reserved.