We are building infrastructure for efficient GPU resource utilization across ML workloads on Kubernetes. You'll work on a production Rust system at the intersection of systems programming, async services, and cloud-native orchestration.
Responsibilities
Async gRPC services and client libraries in Rust
Kubernetes-integrated workload management
Low-level systems integrations (GPU drivers, Linux process model)
Observability: metrics, health endpoints, debug tooling
Integration and end-to-end testing on GPU-backed infrastructure