Deploying Workflows to Production in Dataiku

READ DATASHEET
  • From design to production, instantaneously
  • Robustness & High Availability
  • Feedback loop ready
  • From design to production, instantaneously
  • Robustness & High Availability
  • Feedback loop ready
Product example page showing the graph of a deployed model

From design to production, instantaneously

  • In just a few clicks, deploy your model into production to get real-time predictions with Dataiku DSS' REST API.
  • With bundling of data cleaning, enriching, preprocessing, as well as models for scoring, deploying has never been so fast.
  • Deployed models are versioned, enabling users to rollback anytime.
Product screenshot graphing the predictions of a machine learning model

Scalability & high availability

  • Handle large quantities of real-time predictions with queuing, parallelism, and load balancing.
  • Run multiple scoring API nodes for full high availability.
Product screenshot showing a project bundle

Avoid model drift with a feedback loop

  • Run multiple versions of the same model at the same time for automated A/B testing.
  • Monitor data changes over time.
  • Access history of logs queries and predictions at any time to check that model performance is not drifting with time.