The Challenge: Build the Platform That Powers Research and Beyond

Your mission: Design and deploy the platform that makes GPUs, clusters, and distributed training transparent, not just for internal research, but also as a foundation for monetizable capabilities (e.g., managed training services, optimised inference pipelines for partners).

What You’ll Do

• Cloud & Kubernetes Architecture:

  • Build and maintain a high-performance, multi-tenant environment on Scaleway and GENCI, optimised for distributed ML.

  • Deploy and supervise a Slurm cluster for research workload, ensuring seamless integration with Scaleway’s infrastructure.

  • Automate scaling, resource allocation, and cost management to avoid technical debt.

• MLOps & Internal Tools:

  • Develop and enhance our internal orchestrator to simplify distributed training (FSDP, data pipelines) for both researchers and external users.

  • Create reusable frameworks for monitoring, logging, efficiency, and cost tracking.

  • Collaborate with research teams to industrialise workflows (e.g., model alignment, large-scale finetuning) and package them as deployable capabilities.

• DevOps & Software Craftsmanship:

  • Implement Infrastructure-as-Code (Terraform/Pulumi) for reproducibility and scalability.

  • Write clean, typed, and documented Python code

  • Troubleshoot at the intersection of hardware (GPUs, networking) and software (PyTorch, CUDA), ensuring robustness for both internal and external use cases.

Key Skills

• Experience: 3–4 years in cloud infrastructure, DevOps, or MLOps (research or industry).

• Technologies:

  • Kubernetes/Docker: Advanced orchestration and containerization.

  • Cloud GPU Management: Scaleway, AWS/GCP (clusters, networking, storage).

  • Python: Proficiency in PEP standards, typing, and testing.

  • MLOps: Data pipelines, distributed training (PyTorch, FSDP), monitoring.

  • CI/CD: Pipeline setup and maintenance.

  • Fluent English (the team speaks English in the day-to-day)

Bonus Skills

• Low-level optimisation (Triton, CUDA), HPC, or large-scale training experience.

• Backend/APIs (FastAPI, gRPC) for exposing models or services.

• Experience designing partner-facing tools or managed services.

Beyond Technical Skills:

While technical excellence is critical, we place equal importance on how we work together. We believe the best teams are built on:

• Integrity & Respect

  • We are striving for honesty, kindness, and fairness. We value people who treat others with dignity and foster an environment where everyone feels heard.

• Open Communication & Humility

  • Great ideas come from collaboration. We look for teammates who listen actively, communicate clearly, and approach challenges with self-awareness and humility.

• Psychological Safety & Camaraderie

  • We strive to create a space where people feel safe to take risks, ask questions, and grow.

• Prescreen with Paul (Head of People)

• Technical Screen with one Research Scientist or Research Engineer

• On-site (Take-home exercise and restitution OR On site live interviews + Behavioural interview)