Opening Remarks
summary
AI shifts the infrastructure stack to an AI-first model: designing and custom-building data centers, owning kernel, silicon and software stack, and using PyTorch glue to bind it all. Claims include one of the fastest research supercomputers, the RSC supercluster, custom silicon for inference and video encoding, and a generative AI coding assistant for the SDLC. Facility work uses liquid cooling to shorten construction timelines and cut costs, and six presentations detail these implementations.
AI-written summary · about
related talks
23mSupporting AI holistically Across the Infra stack
25mMaking AI Useful
19mDeveloper Centric AI
06mThe Building Blocks of Agentic Al
13mFaster AI time-to-value when planning
15mKeynote: From Platforms to AI Factories. Has Kubernetes Solved It?
47mEmbedding AI: Don’t Burn Your Fingers in the Heat of the AI revolution
10mTransforming AI into something real