Machine Learning: robust deploy and scaling, sustainable serving, observable production

What is is the first open source platform for Data Science and Machine Learning Management automation. It delivers reliability, scalability and observability for ML and AI applications in production.

Gartner features among Model Management Systems. automates deployment and serving of ML models, monitoring and profiling of production traffic, monitoring of models performance, data subsampling and model retraining.

The platform makes more Data Science and less data plumbing and tinkering happen.

Quick platform overview >


Contact us for a demo


A Case – solutions for AdTech company,
results delivered:

  • Machine Learning operations got scaled from 2 models to 200+ models in production
  • Stabilised and solidified Machine Learning pipelines gave $20M of annual savings.
  • ML Team productivity doubled, estimated ROI increase is $1M per year.
  • Data science production iterations went seamless saving minimum 2 weeks of time per release>.
  • The demand for DevOps people presence in release chain was eliminated completely delivering a solid improvement to ownership costs and ROI.
  • A month of man-hours for product management and a 3 months for QA are saved per release.
  • Apache Spark jobs completion rate reached 99%
  • Cluster throughput increased 10 times saving $100K monthly.
  • Facilitating over 10 products, implementation of the platform into AI/ML operations created a new revenue stream of $10M annually.


Hydrosphere Serving: Realtime Machine Learning Models
Serving Cluster

Deploy your machine learning diversity of sckit-learn, Spark ML, TensorFlow, fastText, xgboost models as end-to-end prediction pipelines. Power smart applications for your users with realtime serving REST API.

  • Quick turnkey endpoint maintenance
  • Serves models trained in any training framework
  • A/B and Canary testing in production
  • Hot model replacement in a production pipeline
  • Robust and simple ML production
  • Get Started > reduces engineering and operations burden and improves time to value metrics for data science projects.


Sonar: Data and Machine Learning QA as a Service

Gain end-to-end quality of your data transformation, training, and prediction pipelines to identify the data quality issues, side effects and model degradation trends before they start affecting your business. provides anomaly detection and pattern recognition components designed for the data and machine learning heavy applications monitoring. It has a great improving impact on the customer experience and the reliability of your data driven business.


Mist: Serverless proxy for Spark cluster jobs management

Make your Spark operations serverless for data scientists, engineers, and multi-tenant applications. increases the reliability of your Spark jobs, thereby saving the cluster resources and increasing the productivity of data scientists and engineers.

  • Exposable REST API for interactive applications to business users and tenants.
  • Parallel Spark jobs execution
  • Fine UI to deploy, test and run jobs
  • Run on AWS:    (?)