Case Study: Hinge Health achieves scalable, reliable, lower-cost data pipelines with Databricks Spark Declarative Pipelines

Developing personalized care plans to improve patient outcomes

Hinge Health, a digital health company helping people reduce musculoskeletal pain, surgeries and opioid use, needed to continuously update personalized care plans as its data grew rapidly. With more than 18 million people served and 1,800 employer health plan clients, the team had to scale data pipelines, cut costs and improve reliability while meeting SLAs. Hinge Health turned to Databricks Spark Declarative Pipelines to simplify change data capture and support its expanding data needs.

Using Databricks Spark Declarative Pipelines, Hinge Health unified batch and streaming workloads, moved from S3 to Kafka as the source, and mirrored source databases into a lakehouse target on Delta. The new architecture improved cluster utilization, reduced intermediary storage and compute costs, and made the pipelines more reliable and easier to scale. Hinge Health now runs 35 data pipelines from a single code base, transforming 300 terabytes of data while better meeting service levels and lowering total cost of ownership, with Databricks.

Hinge Health

Veera Mukkanagoudar

Senior Engineering Manager