Livongo is a leading consumer digital health company on a mission to improve the lives of people dealing with chronic conditions (such as diabetes and hypertension). They offer a whole person platform that empowers members to make smarter decisions. Their suite of connected devices and phone apps process real-world health data to provide actionable, personalized and timely health recommendations.
Livongo is on a mission to use real-world evidence data to provide actionable insights to people at risk for or dealing with chronic health conditions. In order to help their members on their health-care journey, Livongo needs to understand specifics about each member. This requires aggregating large volumes of health data from multiple sources, and then running machine learning models on this data to interpret potential signals about an individual’s health. The signals are then used to suggest healthy actions to members. When Livongo was building out their data architecture they ran into significant hurdles:
Disparate Data Sources: To deliver personalized health recommendations for diabetes patients, Livongo draws on multiple data sets that make it difficult tovefficiently ETL for downstream analytics.
Streamlined Machine Learning at Scale: Delivering health insights in real-time requires fast and reliable machine learning workflows.
Siloed Data Teams: Data teams working on different platforms and in different programming languages caused operational inefficiencies and poor collaboration.
Databricks provides Livongo with a fully managed cloud platform that simplifies operations and allows them to deliver ground-breaking models that improve patient health outcomes.
“Databricks has helped us become more productive quicker by allowing us to reuse a lot of our existing work and improving that with newer models.”