Senior HPC SRE: Scale Research Clusters & Uptime

The Voleon Group; Berkeley (CA); On-site; USD 120,000 - 150,000; Full time

Job summary

The Voleon Group in Berkeley, California is looking for a Senior Cluster Site Reliability Engineer (SRE) to enhance their research compute cluster's reliability and performance. This role requires over 5 years of experience in SRE or DevOps roles, familiarity with HPC frameworks, scripting skills, and cloud infrastructure knowledge. You will ensure maximum uptime, resolve urgent issues, and implement systemic improvements. This is a key position in supporting world-class HPC operations for machine learning research.

Qualifications

• 5+ years of experience in SRE or DevOps roles.
• Knowledge of HPC and machine learning training systems.
• Ability to develop scripts in Python, Ruby, etc.
• Experience with cloud infrastructure (AWS or GCP).

Responsibilities

• Be a first responder in the event of cluster outages.
• Ensure a high degree of cluster uptime and track SLAs.
• Diagnose systemic problems and engineer solutions.
• Develop robust metrics and observability for cluster health.
• Assist in forecasting cluster growth.

Skills

SRE or DevOps experience

HPC/batch compute frameworks

Scripting (Python, Ruby)

Infrastructure-as-code tools

Cloud infrastructure (AWS, GCP)

Modern observability stacks

Distributed storage technologies

System engineer mindset

Education

Bachelor degree in computer science

Tools

Terraform

Ansible

Prometheus

Grafana

Kubernetes

Docker

Slurm

Job description

The Voleon Group in Berkeley, California is looking for a Senior Cluster Site Reliability Engineer (SRE) to enhance their research compute cluster's reliability and performance. This role requires over 5 years of experience in SRE or DevOps roles, familiarity with HPC frameworks, scripting skills, and cloud infrastructure knowledge. You will ensure maximum uptime, resolve urgent issues, and implement systemic improvements. This is a key position in supporting world-class HPC operations for machine learning research.