About the role

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 backed by Y Combinator, Tiger Global, and prominent developer-tool founders, including the founder of Snyk and former CTO of GitHub.

About the Role

We’re looking to hire someone to own inference research hands-on and find ways to lower cost per token and latency on our customer workloads.

• Low-level inference optimization, from speculative decoding, quantization, KV-cache and memory management
• Work directly with customers to optimize their production workloads, and apply your learnings to our platform as product improvements
• High-level of autonomy to find the highest upside bets and guide the future of our inference platform based on your work

Skills & Experience

• Systems or research background in LLM inference
• Deep understanding of LLM serving, from the kernel to the scheduler
• History of shipping products or research that people use in production-like scenarios, whether academic or industry
• Excited to collaborate closely with customers
• Enthusiasm for developer tools, cloud native technologies, and open source software

Benefits

• Competitive salary and meaningful equity
• Join a fast-growing pre-series A company at the ground floor
• Health, dental, and vision benefits with 90% coverage for you and 50% for dependents
• Opportunities to participate in events across the cloud native community
• Fitness stipend, learning budget, and much, much more

About Beam

Cloud computing is broken.

AI has introduced a new generation of workloads, like GPU inference, sandboxes, and agents. These aren't ordinary applications that can be run as Lambdas, or Dockerized apps on VMs: they're massive, stateless containers that need to spin up in <1s, often across multiple clouds and regions.

Today, engineers are hacking together infra that breaks under real-world loads. That's where we come in.

Our mission is to build the world's best compute platform for AI. Our first product is a serverless inference platform, used by companies like Coca Cola, Geospy and hundreds more. We've built our own container runtime, called beta9, which is designed for launching GPU-backed containers in under 1s.

We're a small, highly-technical team, with backgrounds in distributed systems and robotics. We've raised $7M from YC, Tiger, Guy Podjarny (Founder of Snyk), and Jason Warner (former CTO of Github).

We're searching for intensely curious, passionate, and hard-working engineers to join our mission in rebuilding the cloud for the age of AI.