Compute Platform Lead: Multi-Cloud & GPU Systems

Reflection AI Ltd; New York (NY); On-site; USD 180,000 - 300,000; Full time

Benefits offered by this job

Top-tier compensation

Stock options

Health & wellness benefits

Meals provided in office

Parental leave

Job summary

Reflection AI Ltd. is seeking a Compute Platform Lead to guide a multi-cloud, Kubernetes-based compute layer across our neo-cloud footprint.

You will mentor a team of systems engineers, shape architecture for multi-cloud scheduling, cluster management, and next‑generation GPU deployments, while contributing as an IC when required. You will manage vendor relationships, drive scalable, reliable operations, and partner with training teams to ensure fault tolerance and performance.

Qualifications

• Proven experience leading systems or infrastructure teams with hands-on engineering.
• Strong background in cluster-wide behavior, maintenance, and reliability.
• Ability to earn technical trust and guide architectural decisions.

Responsibilities

• Build, mentor, and grow a high-performing systems team.
• Provide front-line leadership for multi-cloud scheduling and cluster management.
• Stay hands-on to contribute technically when needed.
• Prioritize work and manage projects in a fast-paced environment.
• Shape strategy and execution for scalable, robust compute fleets.
• Collaborate with training teams and manage important vendor deals.

Skills

Team leadership

Hands-on systems engineering

Multi-cloud scheduling

GPU deployment expertise

Vendor management

Strategic execution

Tools

Kubernetes

NCCL

Job description

Reflection AI Ltd. is seeking a Compute Platform Lead to guide a multi-cloud, Kubernetes-based compute layer across our neo-cloud footprint.

You will mentor a team of systems engineers, shape architecture for multi-cloud scheduling, cluster management, and next‑generation GPU deployments, while contributing as an IC when required. You will manage vendor relationships, drive scalable, reliable operations, and partner with training teams to ensure fault tolerance and performance.