Uploaded November 2025 | Updated September 2026, 2 weeks ago
At Ray Summit 2025, Jay DeStories and Samuel Jenkins from Tripadvisor share how the company integrated Anyscale into its in-house MLOps platform to unlock the full power of Ray for large-scale AI and ML workloads.
They begin by outlining Tripadvisor’s motivation for adopting Anyscale and the Ray ecosystem—from the need for scalable distributed training and batch inference to improved developer productivity and operational simplicity. The speakers walk through their integration strategy, detailing how they connected Anyscale with existing pipelines, orchestration layers, and internal tooling.
Jay and Samuel then highlight several production ML use cases now powered by Ray, illustrating how the platform handles diverse workloads across recommendation systems, personalization, experimentation, and content understanding. They also share key lessons learned throughout the integration process, including best practices around architecture decisions, reliability patterns, and developer onboarding.
The talk concludes with a deep dive into cost and compute efficiency gains realized as a result of migrating workloads to Anyscale—along with a look at the future roadmap, covering upcoming enhancements, platform expansions, and new AI initiatives Tripadvisor plans to accelerate using Ray.
Attendees will gain a practical perspective on modernizing an MLOps platform with Ray, improving scalability, and driving measurable performance and cost benefits in production.
Liked this video? Check out other Ray Summit breakout session recordings youtube.com/playlist?list=PLzTswPQNepXllnU0C36WtkC0dqkAoDulh
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
At Ray Summit 2025, Jay DeStories and Samuel Jenkins from Tripadvisor share how the company integrated Anyscale into its in-house MLOps platform to unlock the full power of Ray for large-scale AI and ML workloads.
They begin by outlining Tripadvisor’s motivation for adopting Anyscale and the Ray ecosystem—from the need for scalable distributed training and batch inference to improved developer productivity and operational simplicity. The speakers walk through their integration strategy, detailing how they connected Anyscale with existing pipelines, orchestration layers, and internal tooling.
Jay and Samuel then highlight several production ML use cases now powered by Ray, illustrating how the platform handles diverse workloads across recommendation systems, personalization, experimentation, and content understanding. They also share key lessons learned throughout the integration process, including best practices around architecture decisions, reliability patterns, and developer onboarding.
The talk concludes with a deep dive into cost and compute efficiency gains realized as a result of migrating workloads to Anyscale—along with a look at the future roadmap, covering upcoming enhancements, platform expansions, and new AI initiatives Tripadvisor plans to accelerate using Ray.
Attendees will gain a practical perspective on modernizing an MLOps platform with Ray, improving scalability, and driving measurable performance and cost benefits in production.
Liked this video? Check out other Ray Summit breakout session recordings youtube.com/playlist?list=PLzTswPQNepXllnU0C36WtkC0dqkAoDulh
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
