Uploaded September 2026 | Updated November 2026, 2 days ago
AI workloads are evolving fast: data processing, training, and inference jobs run longer, span more accelerators, and lean on more advanced hardware, which makes them more prone to failure. The runtime underneath has to stay ahead.
At Ray Summit 2026, Mengjin Yan, Software Engineer at Anyscale, and Andrew Sy Kim, Software Engineer at Google, cover what Ray Core provides today and how it's evolving along three fronts: reliability, keeping long and expensive runs alive; scalability, pushing the ceiling on the Ray cluster; and hardware, adapting to new accelerators where topology-aware scheduling becomes essential.
You'll leave with a picture of where AI workloads are heading and the work keeping Ray Core ahead of them.
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
🔗 Connect with us:
LinkedIn: linkedin.com/company/joinanyscale
X: https://x.com/anyscalecompute
Website: anyscale.com
AI workloads are evolving fast: data processing, training, and inference jobs run longer, span more accelerators, and lean on more advanced hardware, which makes them more prone to failure. The runtime underneath has to stay ahead.
At Ray Summit 2026, Mengjin Yan, Software Engineer at Anyscale, and Andrew Sy Kim, Software Engineer at Google, cover what Ray Core provides today and how it's evolving along three fronts: reliability, keeping long and expensive runs alive; scalability, pushing the ceiling on the Ray cluster; and hardware, adapting to new accelerators where topology-aware scheduling becomes essential.
You'll leave with a picture of where AI workloads are heading and the work keeping Ray Core ahead of them.
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
🔗 Connect with us:
LinkedIn: linkedin.com/company/joinanyscale
X: https://x.com/anyscalecompute
Website: anyscale.com










