Head of Compute Operations — AI Cloud & Data Centers

Smartshare, Inc.; New York (NY); On-site; USD 180,000 - 240,000; Full time

Benefits offered by this job

Equity

Health benefits

Learning budget

Fitness stipend

Equity plan

Job summary

Beam is seeking a Head of Compute Operations to scale our compute capacity and onboard new partners for our AI platform. You will work with data center operators, OEMs, suppliers, and compute providers to evaluate, test, and contract new capacity.

You’ll manage capacity across datacenter partners, run diligence on new operators, bring clusters online through POC and testing, and build an operations playbook to handle issues, RMAs, and SLAs. Occasional travel to datacenter sites is expected.

Qualifications

• Experience in datacenter ops, capacity planning, or infrastructure delivery at a GPU cloud, neocloud, hyperscaler, or colo provider.
• Able to SSH into a node and discuss InfiniBand vs RoCE.
• Ability to explain how deposits or prepayment schedules affect cash flow.
• Experience managing vendors and holding them accountable when things go wrong.

Responsibilities

• Manage our supply of compute. Keep a live view of capacity across datacenter partners, including hardware type, location, network fabric, operator, and delivery dates.
• Onboard and vet new compute operators. Run diligence and KYC on new partners and review hardware specs and BOMs.
• Bring clusters online. Manage POC and acceptance testing, burn-in, NCCL, interconnect bandwidth; set up bare-metal and BMC access, and hand clusters to customers on the promised date.
• Build our support playbook. Create run books to escalate issues to suppliers, manage RMAs and replacements, and track uptime against SLAs.

Skills

Datacenter ops

Capacity planning

Infrastructure delivery

SSH into a node

InfiniBand vs RoCE

Vendor management

Cash flow awareness

Attention to detail

Tools

NCCL

Job description

Beam is seeking a Head of Compute Operations to scale our compute capacity and onboard new partners for our AI platform. You will work with data center operators, OEMs, suppliers, and compute providers to evaluate, test, and contract new capacity.

You’ll manage capacity across datacenter partners, run diligence on new operators, bring clusters online through POC and testing, and build an operations playbook to handle issues, RMAs, and SLAs. Occasional travel to datacenter sites is expected.