Series

A production ML case study: set-based deep learning that completes Splatoon 3 gear loadouts over a ~160-billion-build space, deployed to real users — problem, architecture, training, and results.

1. 01 SplatGPT: Set-Based Deep Learning for Splatoon Gear Completion 20 min
2. 02 The Deceptive Difficulty of Splatoon 3 Gear Optimization 10 min
3. 03 SplatGPT Part 2A: Engineering GPT for Set Completion 8 min