---
title: Building AlphaGo from scratch – Eric Jang
date: 2026-05-16
slug: alphago-from-scratch-eric-jang
external: https://www.youtube.com/watch?v=X_ZVSPcZhtw
---

[![Building AlphaGo from scratch – Eric Jang](https://i.ytimg.com/vi/X_ZVSPcZhtw/hqdefault.jpg)](https://www.youtube.com/watch?v=X_ZVSPcZhtw)

Eric Jang discusses the process and challenges of building AlphaGo from scratch, sharing insights into deep reinforcement learning, Monte Carlo tree search, and the engineering required to scale up a world-class Go AI. The conversation, hosted by Dwarkesh Patel, covers both the technical and practical aspects of replicating AlphaGo, including:

- The architecture and training pipeline for AlphaGo
- Key breakthroughs in deep RL and search
- Lessons learned from reimplementing complex research systems
- The importance of reproducibility and open science in AI

**Technical Reflection:**

This talk is a must-watch for anyone interested in the intersection of deep learning, game AI, and research engineering. Jang’s experience highlights the value of hands-on replication for understanding state-of-the-art systems, and the discussion offers practical advice for engineers aiming to bridge the gap between academic papers and robust implementations.

[Watch the full interview on YouTube](https://www.youtube.com/watch?v=X_ZVSPcZhtw)


Engineering is the discipline of building things that work within constraints. Every topic on this blog — operating systems, AI models, trading infrastructure, research labs, innovation economics — is examined through the lens of systems design. The lens is engineering. The method is: understand the constraints, design within them, verify the design works, iterate. The domain provides the specifics. The method is universal.