OM1 moved patient-data processing and PII redaction onto Snowflake to cut cost and complexity.
OM1, a Boston healthcare and life sciences data company, ran a costly mix of AWS, Databricks and custom software over billions of rows of patient data, with reliability and scaling problems. It consolidated on Snowpark and Snowpark Container Services for PII redaction, with Modelbit, dbt and Hex for ML deployment and pipelines. The page reports 75% cost savings in processing and redaction of more than 100 million clinical records in under 30 minutes.
Snowflake is a cloud data platform with Cortex AI, used in pharma to unify commercial, clinical and real-world data and to build AI agents on governed data
This sits in the Data Providers stage of the AI use-case lifecycle: Named examples from the real-world-data, claims and competitive-intelligence providers pharma AI systems are trained on, benchmarked against, or run inside of.
Published by Snowflake on its own site — read the original case study on snowflake.com ↗
Related use cases
- 01Sanofi built an AI agent on Snowflake Cortex AI that prepares field reps for every physician visit.
- 02Pfizer uses Snowflake Snowgrid to replicate data across regions for collaboration and business continuity.
- 03Alberta Health Services built an emergency department note-taking app on Snowflake Cortex AI.
- 04How Bioventus evolved their commercial strategy with real world data and intelligence
- 05Clarivate and CLEO diagnostics are making early detection of ovarian cancer a reality
- 06Making informed investment decisions (Attracting breakthrough innovations from academic research) - SPRIND case study