Whitepaper

Generate Business Value with AI Using the Production-First Approach to MLOps

CI/CD for ML

Automate and simplify the building, testing, deployment and monitoring of your ML pipelines and production data. Continuously train, test and deploy your ML models based on changes in your data and business requirements.

Automated, Fast and Consistent CI/CD

Integrate MLRun, Iguazio’s open source framework for orchestrating your ML pipelines, with CI engines to automatically run all ML operations. After each PR and code change, MLRun and the CI engine prepare the data, train models, test, deploy to clusters, monitor the models and send back feedback for retraining and deployment. CI/CD for ML simplifies, streamlines and accelerates the development and deployment process, ensures technological consistency and enables tracking and scalability.

Automated Complex Workflows

Build and run complex workflows composed of local/library functions, external cloud services or other building blocks of your choice.

Tracking and Updating Data and Models

Track and version models, code, data, lineage parameters, artifacts and results with minimal effort.

Integrate with the CI Engine of Your Choice

Run ML pipelines with MLRun and KubeFlow, GitHub Actions, GitLab, Jenkins or other engines.

Ready for Production

Scalable Workloads

Elastically scale each step of the CI/CD pipeline. Specify the configurations, assign GPU, CPU and memory for each workflow step and MLRun will auto-scale.

Data Science For Financial Services
Data Science For Financial Services

Benefits

Automate MLOps

Automate MLOps

Cut Time to Production

Cut Time to Production

Save Resources

Save Resources

Power True Business Impact

Power True Business Impact

Learn More

Data Science Platform Tutorials

Platform Overview

Get started with a video introduction to the Platform

Data Science Platform Documentation

Documentation

Access overviews, tutorials, references and guides

Make Your Data Accessible, Interconnected and Valuable

Learn how to implement the data mesh approach to data architecture with the Iguazio MLOps Platform