A little guide
Zefs Guide to
Deep Learning
The big ideas in deep learning, explained with enough depth to help you understand what's really going on. Available in paperback and ebook, with digital flashcards.
About the guideHi, I'm Roy.
I build and lead teams applying AI to practical problems. My work includes AI strategy and figuring out where it can be useful, building on my experience in machine learning and data science. I still like to build things myself, too.
Based in Houston, USA.
Get in touchA few things I've made
Two books & a benchmarkZefs Guide to
Teams & leadership
Hiring Data Scientists and Machine Learning Engineers
A practical guide to building your team, from planning and hiring to onboarding. Based on my experience and conversations with other hiring managers.
About the bookAn experiment in AI
Code Golf
Benchmark
How small can an AI make a working program? Compare models on code golf challenges in Python, Racket, and Rust.
See the leaderboardSelected writing
All writingThe shortest definition of embeddings?
A compact way to think about learned representations.
Why I don't screen resumes
Rethinking the first step in hiring a data science team.
How I wrote my book
The tools and workflow behind a self-published book.
A little more about me
I'm a physicist by training. My PhD focused on Monte Carlo radiation simulations aimed at improving cancer treatments. Since then, I've built and led data science and machine learning teams at several companies. That experience informs how I approach AI today.
I also co-organize the Houston Data Analytics Meetup and do occasional consulting on strategy and team planning through Zefs Data Science.
You can explore my data projects and experiments, from multi-armed bandits to visualizations.