Uploaded September 2026 | Updated November 2026, 3 hours ago
Grab is transforming its AI platform with Ray to support three ML experimentation loops on one distributed execution layer: offline training, simulator-based parameter search, and reinforcement learning in simulation.
At Ray Summit 2026, Yvonne Wong, Chongyu Zhou, and Abhinav Rai from Grab show through production examples how Ray Train, Data, Tune, and Core simplify heterogeneous scaling, improve experimentation throughput, and make topology, reliability, and cost tradeoffs measurable.
You'll leave with patterns for consolidating experimentation loops on one execution layer.
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
Grab is transforming its AI platform with Ray to support three ML experimentation loops on one distributed execution layer: offline training, simulator-based parameter search, and reinforcement learning in simulation.
At Ray Summit 2026, Yvonne Wong, Chongyu Zhou, and Abhinav Rai from Grab show through production examples how Ray Train, Data, Tune, and Core simplify heterogeneous scaling, improve experimentation throughput, and make topology, reliability, and cost tradeoffs measurable.
You'll leave with patterns for consolidating experimentation loops on one execution layer.
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










