OpsML: Quality Control for the Machine Learning Lifecycle

OSS Version 3.0.0 Coming Soon!

Note from maintainers

Version 3.0.0 is under active development. All pre-releases will be released under the 3.0.0-rc.* tag.

OpsML Documentation

Task Backlog

What is it?

OpsML is a developer-first ML operations platform focused on injecting quality control into the machine learning lifecycle. Through automation and standardization, OpsML provides a unified interface and experience for managing ML artifacts, enabling teams to collaborate more effectively and deploy with confidence, all while reducing engineering overhead and providing peace of mind.

What is Quality Control?

Quality control in the context of OpsML refers to:

Developer-First Experience

• Zero-friction Integration - Drop into existing ML workflows in minutes
• Type-safe by Design - Rust in the back, python in the front^*. Catch errors before they hit production
• Unified API - One consistent interface for all ML frameworks
• Environment Parity - Same experience from laptop to production
• Dependency Overhead - One dependency for all ML artifact management

Built to Scale

• Trading Cards for ML - Manage ML artifacts like trading cards - collect, organize, share
• Cloud-Ready - Native support for AWS, GCP, Azure
• Modular Design - Use what you need, leave what you don't

Production Ready

• High-Performance Server - Built in Rust for speed, reliability, and concurrency
• Built-in Security - Authentication and encryption out of the box
• Audit-Ready - Complete artifact lineage and versioning
• Standardized Governance - Consistent patterns across teams
• Built-in Monitoring - Integrated with Scouter

Us vs Others

Contributing

If you'd like to contribute, be sure to check out our contributing guide! If you'd like to work on any outstanding items, check out the roadmap section in the docs and get started.

Thanks goes to these phenomenal projects and people for creating a great foundation to build from!