GPU & Kubernetes SRE Engineer for AI Infrastructure

Together Computer Inc; United States; Remote; USD 140.000 - 190.000; Vollzeit

Zusammenfassung

Together AI is seeking a customer-facing Site Reliability Engineer to manage Kubernetes GPU clusters and ensure high availability. The role involves monitoring GPU health, addressing hardware issues such as thermal throttling and NVLink degradation, and maintaining a Slurm-based production cluster.

You will work with Engineering and Product teams to advance the roadmap. Ideal candidates have 3+ years in customer-facing tech roles and 1+ year supporting AI services or critical SaaS APIs, with

Qualifikationen

• 3+ years in a customer-facing technical role.
• 1+ year supporting AI services or mission-critical SaaS APIs.
• Experience as an SRE or DevOps engineer with Kubernetes.
• Knowledge of AI, ML, and GPU technologies in HPC environments.
• Proficiency with infrastructure services like SLURM and Ansible.
• Experience with high-speed networking such as InfiniBand and RDMA.
• Experience with distributed storage systems like Weka or NFS.
• Strong troubleshooting skills for compute clusters and I/O issues.

Aufgaben

• Resolve complex technical challenges involving Kubernetes GPU clusters.
• Act as a customer-facing SRE to ensure cluster health and stability.
• Monitor GPU cluster health and communicate hardware issues like thermal throttling or NVLink degradation.
• Maintain production infrastructure including Slurm cluster maintenance and node repair.
• Investigate storage and networking issues involving Weka filesystem or InfiniBand.
• Collaborate with Engineering and Product teams to drive the product roadmap.
• Maintain detailed documentation of system configurations and troubleshooting guides.

Kenntnisse

Customer-facing role

Kubernetes SRE/DevOps

AI/ML HPC knowledge

SLURM

Ansible

InfiniBand/RDMA networking

Weka/NFS storage

Cluster troubleshooting

Tools

Kubernetes

SLURM

Ansible

Jobbeschreibung

Together AI is seeking a customer-facing Site Reliability Engineer to manage Kubernetes GPU clusters and ensure high availability. The role involves monitoring GPU health, addressing hardware issues such as thermal throttling and NVLink degradation, and maintaining a Slurm-based production cluster.

You will work with Engineering and Product teams to advance the roadmap. Ideal candidates have 3+ years in customer-facing tech roles and 1+ year supporting AI services or critical SaaS APIs, with