Back to jobs
Director, Engineering, AI Accelerator Cluster Systems
Posted 2 weeks ago
Job Description
- Build physical infrastructure and custom networking, while designing data centers focused on custom cooling, rack density, and CDU placement.
- Oversee the core bare-metal software stack, including drivers, OS, firmware, and accelerator release management.
- Engineer systems with direct GPU/TPU access, ensuring high compute density and low latency for training foundation models.
- Partner cross-functionally to optimize the end-to-end AI accelerator stack from large-scale networking to Kubernetes.
- Collaborate with product and GTM leaders to shape the multi-year bare-metal AI infrastructure strategy.