Building Self-Serve, Multimodal Medical Image Pipelines with Ray | Heartflow | Ray Summit 2026 @anyscale
Building Self-Serve, Multimodal Medical Image Pipelines with Ray | Heartflow | Ray Summit 2026  @anyscale
Uploaded September 2026 | Updated November 2026, 3 days ago
Heartflow's internal data platform turns data transformations into fault-tolerant, scalable pipelines running nightly on Anyscale, with no Ray code required.

At Ray Summit 2026, Deepak Sharma and Patrick Violette from Heartflow show how the platform, built on Ray Data and Ray Core, provides self-serve medallion pipelines that autoscale on spot instances, evolve schemas, and publish to Iceberg, powering everything from coronary plaque analytics to metadata extraction from millions of medical images.

You'll leave with their architecture, how they scaled it, and lessons from running production Ray across teams.

Liked this video? Check out other Ray Summit breakout session recordings: youtube.com/playlist?list=PLZNaJyYZRPdo

Subscribe to our YouTube channel to stay up-to-date on the future of AI: youtube.com/c/anyscale

đź”— Connect with us:
LinkedIn: linkedin.com/company/joinanyscale
X: https://x.com/anyscalecompute
Website: anyscale.com
Building Self-Serve, Multimodal Medical Image Pipelines with Ray | Heartflow | Ray Summit 2026Agents and Robots: The Human Effort Putting AI Into Production | Encord | Ray Summit 2026How Exa Indexes the Entire Web: Ray Data and Lance at Billion-Document Scale | Exa | Ray Summit 2026Unlocking Peak Workload Performance & Efficiency with Ray on Kubernetes | Ray Summit 2025Boosting vLLM Inference on Huawei NPU with Ray Compiled Graphs — Huawei | Ray Summit 2025Simulating the Physical World at Scale: Humanoid Locomotion with Ray | Lambda | Ray Summit 2026KV Cache and Long-Context Inference: Solutions for Inference at Scale | VAST Data | Ray Summit 2026Reward Hacking and the Gaps in Agentic AI | Ion Stoica (Anyscale) | Ray Summit 2026How AWS Scales Reinforcement Learning Across Thousands of GPUs | Ray Summit 2025Distributed Model Training with Ray at Capital One | Ray Summit 2025State of vLLM 2026 | Inferact | Ray Summit 2026From Benchmark #1 to Production | Nebius | Ray Summit 2026
Anyscale |

Building Self-Serve, Multimodal Medical Image Pipelines with Ray | Heartflow | Ray Summit 2026

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER