jobs
Senior GPU Performance Software Engineer, AI/ML
Google's software engineers develop the next-generation technologies that change how billions of users connect, explore, and interact with information and one another.
En inglés
- Company
- Location
- London, UK
- Status
- Closed
- Posted
- 2026-10-08 10:12 UTC
Google's ML-Omega team optimizes, models and benchmarks GPU systems for internal machine-learning workloads. The engineer maintains LLM training and serving benchmarks that reflect production, using them to find performance wins and guide XLA GPU and Triton work; engages product teams including DeepMind on onboarding and scaling large models; runs architecture-level GPU simulations and roofline analysis; and hunts fleet-wide bottlenecks. The posting asks for five years of software development, data structures and algorithms, and machine-learning experience, plus GPU programming in CUDA or Triton. Preferred: hardware/compiler co-design, HPC, distributed performance work.
Original job posting