In this talk we will share how we use Ray to address various challenges Samsara faced building its machine learning platform from scratch, specifically, creating a unify but flexible development experience for a team that needs to build AI in both cloud and edge.
We will deep dive on how Ray and its ecosystem (Spark on Ray, Dagster etc.) enables ML research and development on a non-traditional setup, which improves data exploration and model training experience. We will also share our experience on managing our Ray infrastructure and some of the best practices we learn along the way.
I lead machine learning infrastructure initiatives in Samsara, currently working with Ray and many other OSS communities to deliver a scientist centric experience in Samsara's machine learning platform.
Saurabh Tripathi is a senior machine learning engineer at Samsara’s ML Infrastructure team. He has helped in setting up the Orchestration and Model Monitoring framework at Samsara using Ray and other Open source tools.
Sharan Srinivasan is an Applied Scientist at Samsara. He works on problems relating to computer vision for fleet and industrial safety applications - focusing on driver safety and ADAS systems. Previously he was part of the Search and Market Dynamics team at Airbnb He holds an MS from Stanford University, where he focused on Operations Research.
Come connect with the global community of thinkers and disruptors who are building and deploying the next generation of AI and ML applications.