Overton: A Data System for Monitoring and Improving Machine-Learned Products

TL;DR

Overton is a system designed to assist engineers in building, monitoring, and improving production machine learning systems through high-level abstractions, automating lifecycle tasks, and enabling deep learning applications without coding.

Contribution

It introduces novel high-level abstractions for ML system management, allowing engineers to build and monitor ML applications more efficiently and without extensive coding.

Findings

Overton has been used in production for over a year.

Applications using Overton answered billions of queries.

Overton reduced errors by 1.7-2.9 times compared to existing systems.

Abstract

We describe a system called Overton, whose main design goal is to support engineers in building, monitoring, and improving production machine learning systems. Key challenges engineers face are monitoring fine-grained quality, diagnosing errors in sophisticated applications, and handling contradictory or incomplete supervision data. Overton automates the life cycle of model construction, deployment, and monitoring by providing a set of novel high-level, declarative abstractions. Overton's vision is to shift developers to these higher-level tasks instead of lower-level machine learning tasks. In fact, using Overton, engineers can build deep-learning-based applications without writing any code in frameworks like TensorFlow. For over a year, Overton has been used in production to support multiple applications in both near-real-time applications and back-of-house processing. In that time, Overton-based applications have answered billions of queries in multiple languages and processed trillions of records reducing errors 1.7-2.9 times versus production systems.