Senior Software Engineer, Infrastructure ($160k-$250k + Equity) at well-funded AI video platform

About this role

You will own the GPU inference infrastructure powering real-time AI conversations with sub-second latency. Managing multi-provider GPU deployments and EKS clusters, you'll optimize performance from CUDA hot paths to backend services. This role is critical for scaling a platform that enables seamless human-to-AI interaction across global regions.

Why this role is remarkable

• Own the core infrastructure for a cutting-edge real-time AI video interface that operates at sub-second latency levels globally.
• Join a well-funded Series B startup backed by top-tier VCs, operating with a small, high-impact team of roughly 40 people.
• Work at the intersection of infrastructure and research, directly influencing how conversational AI humans are deployed and experienced.

What you will do

• Design and manage multi-region GPU deployments across various providers to ensure reliable, high-performance inference.
• Architect EKS clusters including custom routing and scheduling logic to optimize workload placement across a global GPU fleet.
• Collaborate with researchers to optimize CUDA hot paths and reduce cold-start times for live, face-to-face AI conversations.

What we're looking for

• Hands-on experience deploying and optimizing large-scale GPU inference workloads on cloud providers like CoreWeave or AWS.
• Deep expertise in Kubernetes and EKS, specifically regarding complex service routing and cluster management at scale.
• Proven track record of senior technical leadership, taking ownership of ambiguous infrastructure challenges and delivering production-ready solutions.

Compensation

Salary range: $160k–$250k + Equity

Location

San Francisco, USA or Remote