Beam

Beam is an ultrafast AI inference platform. We built a serverless runtime that launches GPU-backed containers in less than 1 second and quickly scales out to thousands of GPUs. Developers use our platform to serve apps to millions of users around the globe. We're working on challenging problems, including low-level systems development, efficiently packing thousands of workloads into GPUs across multiple clouds, and working with cutting-edge technologies like GPU checkpoint restore and CRIU. Beam is a tool to quickly build machine learning-powered applications. Our platform helps developers run their code on serverless GPUs, deploy highly performant APIs, and rapidly prototype ML models — without managing any infrastructure. Machine learning is eating software, but it’s still difficult for developers to leverage ML in their products. Today, companies are spending months building their own ML platforms, or relying on outdated tools that were originally designed for academics. We believe that for ML to reach widespread adoption, the underlying infrastructure needs to be hidden from the user. We're building the fastest way for developers to go from an ML prototype to a production service.