Staff Software Engineer (Compute Efficiency), London
Core
Design, deploy, and scale observability systems and telemetry pipelines to monitor fleetwide compute efficiency, hardware health, and workload goodput across distributed GPU clusters.
Role type
Staff Software Engineer (Compute Infrastructure & Observability)
Builds
Observability systems, automated recovery mechanisms, and telemetry pipelines for planet-scale accelerator fleets.
Domain
Biotech / AI Infrastructure / Cloud Compute
Deliverable
infrastructure
Required skills
Large-scale AI/ML workload infrastructure operations, Cloud compute infrastructure design (GCP), Kubernetes at scale, Programming, Nvidia GPU familiarity, Production observability and telemetry stack development
Preferred skills
ML SWE or Infrastructure SRE background, ML workload efficiency paradigms, Project leadership, Google TPU familiarity, Workload scheduling, ML-driven R&D cycles, Hardware benchmarking
Technologies
GCP, Kubernetes, Nvidia GPUs, Google TPUs
Responsibilities
Design and deploy robust observability systems and telemetry pipelines; Drive hardware efficiency and node reliability across the accelerator fleet; Identify compute waste and efficiency bottlenecks; Collaborate with AI and ML Infrastructure teams to improve canonical efficiency metrics; Operate, maintain, and harden research, development, and production cloud infrastructure; Contribute to core technical decisions regarding tooling and architecture.
