Our backend is mostly Go and Python — Go for the integration layer and high-throughput services, Python for the ML pipelines and clinical orchestration. Postgres is our primary store, with ClickHouse for analytics and Kafka for event streaming.
The frontend is React and TypeScript. We deploy on AWS, with a small footprint in GCP for specific ML workloads.
We choose technologies we can operate, not technologies we can put on a slide.
