ML Engineer – Infrastructure

About the role

Flexion Robotics is building the autonomy stack for humanoid robots and needs a senior ML Engineer to own its GPU compute platforms. You will design, bring‑up, operate and optimise large‑scale GPU clusters that train foundation models for humanoid AI.

Key responsibilities

• Architect, provision, and continuously improve cloud‑based GPU clusters, selecting frameworks and tooling for efficiency.
• Support AI engineers by optimising training workloads, using profilers and contributing to core ML libraries.
• Develop short‑ and long‑term GPU compute strategies, explore multi‑cloud options and evaluate trade‑offs.
• Improve capacity and cost efficiency, and raise engineering standards for testing, documentation and reliability.

Required profile

• Degree in Computer Science, Electrical Engineering, Software Engineering or equivalent practical experience.
• Significant industry experience with training or inference of large models (billions of parameters) on distributed multi‑node GPU hardware.
• Hands‑on experience with cloud platforms (AWS, GCP, Azure) or large‑scale on‑premises GPU infrastructure.

Required skills

• Python programming.
• PyTorch and deep understanding of distributed training (DDP, FSDP, NCCL).
• Job scheduling/orchestration tools such as Slurm and/or Kubernetes/KubeRay.
• Experience with infrastructure‑as‑code tools (Terraform, Ansible) is a plus.

What we offer

• Competitive compensation and equity.
• Access to cutting‑edge robotics research and testing facilities.
• Energetic, collaborative culture with regular community events.
• Enhanced pension plan, relocation & permit sponsorship, and generous holiday benefits.
• Health, dental & vision coverage with flexible plan options.