The requirements for building, deploying and managing AI applications in production mean significant MLOps and engineering efforts, rendering the old training-first paradigm insufficient. The solution? AutoMLOps.
AutoMLOps means automating the many engineering tasks of deploying ML, so that your code is automatically ready for production. Oh, and there are open-source tools out there that enable it! AutoMLOps includes:
- Automatically converting code to managed microservices and reusable components
- Auto-tracking experiments, metrics, artifacts, data, models
- Automatically registering models along with their required metadata and optimal production formats
- Auto-scaling and automatically optimizing resource usage (such as CPUs / GPUs)
- Codeless Integration with different dashboards, profilers, CI/CD frameworks, etc.
In this session, we outline the challenges, describe open-source tools available for Auto-MLOps, and finish off with a live demo.