lrakla/ML-in-Production

It is easy to prototype ML models. With higher levels of abstraction, the expertise required to build any ML model is decreasing each day.

ML-in-Production

It is easy to prototype ML models. With higher levels of abstraction, the expertise required to build any ML model decreases daily. It is not easy to make an ML model available to millions of users, maintain it, and monitor it. In this repo, I want to consolidate resources to help people take their models to production, document what leading tech companies do, and how to get started with MLOps.

1. Rules of ML
2. 150 Successful Machine Learning Models: 6 Lessons Learned at Booking.com
3. Hidden Technical Debt in Machine Learning Systems