Technical Program Manager

The role

You will own provider relationships on the execution side by running programs for onboarding, capacity rollouts, and incident response. You will act as an incident commander for provider-side failures and define the provider program management process.

What you will do

• Manage new provider and site onboarding, capacity expansions, and remediation of hardware or network issues.
• Act as incident commander during major provider-side incidents.
• Maintain plans for each program including milestones, owners, dependencies, and risks.
• Hold providers to contractual commitments through check-ins and escalations.
• Coordinate internally with SRE, Engineering, Product, and Sales/Customer Success.
• Identify capacity and quality risks early.
• Create playbooks and provider-facing standards.

Required qualifications

• Several years running technical programs in infrastructure.
• TPM or technical project management with execution ownership involving external vendors, partners, or suppliers.
• Incident management experience commanding or running point on production incidents involving multiple organizations.
• Comfort with off-hours reality of incident management.
• Technical depth regarding GPUs, networking, and storage.
• Track record of aligning teams you do not manage, including external partners.
• Skills in planning, risk tracking, and status communication.
• Calm, direct communication.
• Comfort with ambiguity.

Nice to have

• Direct experience working with or inside neocloud, colocation, or data-center providers.
• Vendor or supplier management background — SLAs, commitments, escalation frameworks.
• Familiarity with the stack: NVIDIA data-center GPUs, InfiniBand/RoCE, Slurm or Kubernetes.
• Experience supporting AI research labs or other large-scale GPU customers on the consuming side.

Benefits

• Healthcare, dental, and vision coverage for you and your dependents
• 401(k)
• Unlimited PTO

About the company

Andromeda is a market and infrastructure platform to buy, sell, and operate compute. The company aims to enable fragmented compute to flow through one platform to serve model builders, research labs, and inference providers.