🛠 Tech Stack

💼 About This Role

You'll oversee the high-performance computing environment powering frontier AI research at a non-profit institute. Your work will directly enable breakthroughs shared with the global scientific community, balancing on-prem GPU clusters with hybrid cloud orchestration. You'll shape the infrastructure strategy for the next 3–5 years of AI research.

🎯 What You'll Do

• Manage on-prem GPU cluster availability and vendor partnerships
• Optimize job scheduling on internal orchestration platform (Beaker)
• Develop storage roadmap balancing throughput and cost durability
• Make data-driven decisions on cloud vs. on-prem compute investment

📋 Requirements

• 12+ years in infrastructure or HPC with 5+ years leadership
• Bachelor’s degree in related field (advanced degree may substitute)
• Hands-on experience with NVIDIA GPU clusters and InfiniBand/RoCE
• Proficiency in Kubernetes or Slurm for hybrid-cloud orchestration

✨ Nice to Have

• Experience with distributed filesystems (WEKA, Ceph, Lustre)
• Proficiency in Go or Python for internal tooling development

🎁 Benefits & Perks

• 💰 Competitive base salary plus generous bonus plans
• 📚 On-going education via Ai2 Academy and guest speakers
• ⚖️ Healthy work/life balance with generous paid vacation and sick leave
• 🤝 Collaborative culture with transparency and shared problem-solving
• 🏢 Seattle office on the water with scenic location

🚩 Heads Up

• On-site requirement with limited remote flexibility
• Requires 12+ years experience with 5+ years leadership, which may be excessive for title
• Physical demands clause (stationary for long periods) listed under essential functions