case study

ML Ops Foundational Framework for Data Science and Analytics

Insurance Snowflake

Business Impacts


faster speed of end to end cycle

5 million+

predictions generated per month


time saved for model deployment


time saved for ingestion

Customer Key Facts

  • Country : North America
  • Industry : Insurance
  • About : One of the foremost insurance corporations in America, this client stands as the largest provider of supplemental insurance across the nation. Renowned for their comprehensive range of products, they lead the market in delivering specialized insurance solutions that cater to a diverse clientele.

Problem Context

The client’s data was on-prem, limiting speed and flexibility for compute scaling. They needed a ML framework for data preparation, pre & post processing operations, one-click automated ML model training and inference pipelines, but lacked support for a data platform like Snowflake, limiting data science and BI capabilities


  • Data duplication & redundancies while accessing updated data
  • Scaling of compute resources is challenging
  • No data democratization & lack of ability to generate data driven insights
  • Poor data governance due to lack of visibility into PII/PHI data being used across different models

Technologies Used











Data Robot

Data Robot


Client’s data was collected from various source systems and ingested into Snowflake to provide a single source of truth. Snowflake helped in providing the base layer to support the ML Framework for automated training, inference pipelines and model management.

Model Outputs Stored in Snowflake storage table. Implementing Snowflake also helped in building custom dashboards and provided an Interface to view results and generate insights.


  • We have achieved a remarkable acceleration in the speed of the end-to-end cycle.
  • Reached an impressive capacity for generating a substantial volume of predictions monthly.
  • We have significantly reduced the time required for model deployment.
  • We have drastically cut down the time needed for data ingestion

Thank you for reaching out to us!

Our experts will be in touch with you shortly.

In the meantime, explore our insightful blogs and case studies.

Something went wrong!

Please try it again.