Remote Technical Lead for GPU Infrastructure & Slurm

Tether; Dublin; On-site; EUR 120,000 - 180,000; Full time

Benefits offered by this job

Remote work

Job summary

Tether, a global fintech leader, is seeking a Technical Lead for Cosmic AC—a GPU compute and managed inference platform. You will own architecture end to end, lead an international engineering team, and drive a hybrid of bare-metal GPU scheduling with Kubernetes control plane for high‑scale inference.

You will partner with research and product teams to translate workloads into scalable platform requirements, ensure reliability, and guide capacity planning.

Qualifications

• Eight+ years hands-on engineering experience with infrastructure platforms.
• Lead teams that build and operate multi‑tenant platforms.
• Bachelor's or Master's in CS/engineering or equivalent experience.

Responsibilities

• Own platform architecture end to end with designs and reviews.
• Lead and line-manage a distributed team across backend, frontend, DevOps, QA and docs.
• Design and operate a managed Slurm service for research workloads.
• Own Kubernetes control plane and GPU enablement for bare metal clusters.
• Oversee scalable inference architectures and multi-GPU deployments.
• Maintain observability, SLIs/SLOs, incident response and on-call readiness.
• Interface with partners/vendors to translate requirements into tests and specs.
• Collaborate with research/model-training teams to translate workloads into platform needs.

Skills

Leadership

Architecture decisions

JavaScript/Node.js fluency

Team management across time zones

Cloud and Kubernetes experience

Engineer collaboration with research &

Education

Bachelor's or Master's in CS or Engineering

Tools

Node.js

React

Kubernetes

Slurm

NVIDIA GPU Driver/CUDA lifecycle

Docker

Job description

Tether, a global fintech leader, is seeking a Technical Lead for Cosmic AC—a GPU compute and managed inference platform. You will own architecture end to end, lead an international engineering team, and drive a hybrid of bare-metal GPU scheduling with Kubernetes control plane for high‑scale inference.

You will partner with research and product teams to translate workloads into scalable platform requirements, ensure reliability, and guide capacity planning.