Uploaded September 2026 | Updated November 2026, 2 days ago
Fragmented multimodal training causes high latency and low GPU utilization.
At Ray Summit 2026, Jeff Braunstein, Senior Director of Product at CoreWeave, presents a unified architecture that eliminates those silos: Anyscale's managed Ray running natively on CoreWeave Kubernetes Service with CoreWeave AI Object Storage, streamlining video curation at petabyte scale, with benchmarks showing gains in throughput, utilization, and TCO.
You'll leave with a reusable reference architecture for efficient, integrated training.
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
Fragmented multimodal training causes high latency and low GPU utilization.
At Ray Summit 2026, Jeff Braunstein, Senior Director of Product at CoreWeave, presents a unified architecture that eliminates those silos: Anyscale's managed Ray running natively on CoreWeave Kubernetes Service with CoreWeave AI Object Storage, streamlining video curation at petabyte scale, with benchmarks showing gains in throughput, utilization, and TCO.
You'll leave with a reusable reference architecture for efficient, integrated training.
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










