AI Platform Engineer

Job Description

The Advanced Scientific Compute (ASC) team is evolving. As we shift our focus toward hosting proprietary models and managing state-of-the-art, on-premise GPU hardware for both AI model training and inference, we are looking for a motivated AI Platform Engineer.

In the modern AI landscape, we are looking for an engineer with a strong technical foundation and good understanding of AI concepts, who can also act as an orchestrator. You will help operate and support the foundational layer of our on-premise AI hosting environments—encompassing GPU/HPC infrastructure, containerization, automation, and observability. You will leverage AI agents to augment your technical skills, accelerating your workflows and allowing you to focus on problem-solving and collaboration with our research community across the US and EU.

• Manage and automate on-premise infrastructure for AI and HPC, focusing on high-performance GPU servers and clusters, while occasionally assisting with cloud-based resources.
• Deploy, manage, and orchestrate AI training and inference workloads, primarily utilizing Slurm, Nvidia Triton and Docker containers.
• Leverage AI and GenAI tooling to accelerate development, build CI/CD pipelines, and automate repetitive infrastructure tasks.
• Assist in performance tuning, capacity tracking, and scaling of compute clusters to ensure reliable and efficient services for researchers.
• Collaborate directly with researchers to understand workflows and translate non-technical needs into robust infrastructure solutions.
• Maintain monitoring, dashboards, logging, and alerting for critical components to support vital AI infrastructure.
• Maintain and evolve the existing ecosystem of HPC modules and scientific applications to ensure reproducible builds and consistent environments.
• Share expertise in leveraging agentic AI tools and champion modern, AI-assisted engineering practices among team members.

• 3+ years of software or infrastructure engineering experience, with at least 1 year of hands-on experience working with AI platforms, GPU environments, or similar modern compute workloads.
• Practical experience in Enterprise Linux administration, automated testing, CI/CD pipelines, and Infrastructure as Code tools (e.g., Terraform, Ansible).
• Strong experience with Docker and container runtimes, alongside familiarity with job schedulers like Slurm in an HPC or scientific computing environment.
• Practical experience with GPU servers or clusters, including a solid understanding of NVIDIA drivers, CUDA/cuDNN/NCCL, and the NVIDIA Container Toolkit.
• Solid Python programming skills and strong DevOps fundamentals, particularly utilizing version control (Git).
• Experience with monitoring and troubleshooting distributed systems or compute clusters.
• Strong communication skills and the ability to work flexibly to collaborate across EU and US timezones.

• Experience with Kubernetes and operating containerized platforms at scale.
• Background in pharma, biotech, or academic scientific computing environments.
• Full-stack development capabilities (web UIs, REST APIs) to build internal researcher-facing tooling.
• Familiarity with Python packaging and environment management tools (Conda/mamba, pip) alongside the scientific Python stack.

Current Employees apply HERE

Current Contingent Workers apply HERE