Join a rare convergence of Google's most distinguished engineers and Fellows for a candid discussion on the next frontier of artificial intelligence. This panel delves into the critical hardware, software, open-source principles, and systems-level innovations that are solving the unique challenges of AI inference at scale. Hear directly from the minds who shaped Google's foundational technologies as they explore the history and future of inference, the role of compilers like XLA in bridging models and silicon, and the commitment to open source and sustainability that drives a more efficient and powerful AI ecosystem. This session offers a unique look into the strategic thinking and deep technical investments that are paving the way for the next decade of AI.

Go behind the scenes with the technology leaders from innovative companies in this technical panel discussion.  We’ll explore real-world applications of cloud infrastructure for cutting-edge AI workloads. Companies will showcase their AI infrastructure learnings, and discuss the critical technical trade-offs and architectural decisions made to support and scale demanding AI workloads.

As AI models transition from training to production, the critical challenge becomes delivering fast, reliable, and cost-effective inference at scale. 

In this exclusive, private session, we will share our strategic roadmap for Google Cloud's inference-optimized infrastructure. Get an inside look at how we are engineering our hardware and software to tackle the unique demands of large-scale AI workloads. We will detail upcoming advancements designed to deliver ultra-low latency, maximize throughput, and provide breakthrough price-performance for your most critical inference workloads. This is an interactive discussion where you can ask questions and provide direct feedback to the product leaders.