Deskripsi Pekerjaan
HyperScale AI is seeking an exceptional Senior AI Infrastructure Engineer to design, deploy, and scale our high-throughput distributed training platforms. You will manage large-scale multi-node GPU clusters, eliminating hardware-software bottlenecks for cutting-edge foundation models.
We provide comprehensive health benefits, significant equity upside, and modern hybrid office spaces in downtown San Francisco.
Tanggung Jawab
- Architect and maintain high-performance GPU clusters for distributed machine learning workloads.
- Optimize large-scale data ingestion and model checkpointing pipelines across heterogeneous environments.
- Automate infrastructure provisioning, configuration, and monitoring with Kubernetes and Terraform.
- Collaborate with ML research scientists to debug CUDA and distributed framework bottlenecks.
- Implement automated cluster recovery, node health monitoring, and zero-downtime deployment pipelines.
- Enforce security controls, role-based access policies, and isolation patterns across multi-tenant environments.
Kualifikasi
- Bachelor’s or Master’s degree in Computer Science or equivalent practical experience.
- 5+ years of production experience in backend systems or site reliability engineering.
- 3+ years of dedicated experience running workloads on Kubernetes and large GPU clusters.
- Proficiency in Python, Go, and low-level Linux performance tuning.
- Hands-on experience with distributed training tools such as Ray, DeepSpeed, or Slurm.
- Strong background in network optimization including RoCE, InfiniBand, and cloud VPC peering.