Machine Learning Systems Engineer in charge of GPU Infrastructure 50 - 100%

  • ZHAW Zurich University of Applied Sciences
  • Winterthur, ZH, Switzerland
  • 01/05/2023
Full time Data Engineering Machine Learning Business Intelligence Software Engineering

Job Description

Job details

No matching job preferences
Job Type

Full Job Description

You plan and operate our growing critical GPU computing infrastructure and act as interface to researchers and IT (

School: School of Engineering
Starting date: June 2023 or by appointment

Your role

  • Being the go-to person at the CAI for all aspects of our mission-critical on-premise or cloud-based GPU infrastructure (see below) in a highly dynamic environment that currently includes planning, procurement, building, and operations; in the future, lower-level tasks might be handled by central IT
  • Collaborating with the head of the CAI demo team (who oversees other CAI infrastructure) and interfacing with the researchers and students to continually adapt the GPU infrastructure to user needs, provide training and best practices, etc.
  • If employed over 50%: participation in applied research and development projects as ML engineer / researcher in the IVS research group (deep learning, model optimization, MLOps pipelines, publications, etc.)
  • Examples of existing compute infrastructure (on-premise): 4 NVIDIA DGX servers; a virtualization environment with GPUs and Ceph storage; cluster management with Slurm and Docker

Your profile

  • We are looking for a pragmatic person with a strong drive to build things that work and are appealing to use; we expect a strong background in scientific computing, machine learning
  • deep learning, and programming, with a master's or PhD degree in computer science or a comparable background
  • You have proven competencies in the areas of ML engineering (e.g., scikit-learn, PyTorch, CUDA, Kubernetes, MLOps, i.e., development, operation, and maintenance of ML models and pipelines) and server administration (Linux, storage organization, hardware setup for ML). Experience with cloud-based GPU computing providers is a plus.
  • Your curiosity makes you maintain an overview on the current state of knowledge in applied AI and related technology, to be considered for additional tasks in research, you have a strong AI-related research track record (e.g., proven by publications), ideally related to computer vision and MLOps.
  • Fluency in English and ideally German as well as the capacity to interact on equal footing with management, researchers and students is important; you work independently, manage projects in autonomy, enjoy cooperation in interdisciplinary teams, and have an accurate, result-oriented working style.
  • In return, we offer a position on the level of a research associate, initially limited to 2 years, in which you ideally just live out your hobby (playing with latest technology), freedom and development opportunities (for example in the direction of research, project management or an expert role for MLOps), as well as a high degree of flexibility with respect to working times and places (up to 50% homeoffice)