Key Responsibilities:
- High Performance Computing (HPC) Specialist for Novartis Core Infrastructure Services team performs the operational health deployment and lifecycle of HPC technology
- Designs innovative solutions utilizing projects and qualification frameworks to run and optimize our complex infrastructures.
- Act as technical and organizational escalation point during major and critical incidents.
- Responsible to track suppliers and partner for effective and efficient delivery of projects
- Contributes to service / platform strategy development
Commitment to Diversity & Inclusion:
We are committed to building an outstanding, inclusive work environment and diverse teams representative of the patients and communities we serve.
What you ll bring to the role:
- Associates degree in computer science or related field, or equivalent industry experience and 6+ years IT experience, of which at least 3 in an HPC/scientific computing environment. Worked in environments with high speed / low latency network (e.g. InfiniBand)
- Experience designing and building infrastructure within a public cloud platform (e.g. AWS, GCP or Azure) as they relate to HPC workloads. Strong knowledge of Unix systems performance tuning. Working knowledge of DevOps (e.g. Ansible, Chef, CICD etc.) tools and practices
- Hands-on experience with system administrative task in Linux environment, and fluency in scripting (shell scripting in bash required, Python knowledge would be advantageous)
- Can build software from sources including knowledge of build systems (e.g. Make, CMake...). Extensive knowledge of one or more HPC scheduling mechanisms (e.g. Grid Engine, Slurm, LSF... etc.). Experience with shared memory and/or GPU and/or distributed memory computation jobs
- Extensive knowledge of one or more HPC cluster management software packages (e.g. Bright Cluster Manager / Base Command Manager, xCat, OpenHPC etc.). Solid knowledge of computer architectures and multi-threaded/parallel processing applications. Hands-on knowledge of network- and distributed filesystems (e.g. NFS, GPFS, Gluster, BeeGFS, Lustre or other parallel file systems, etc.)
Desirable requirements:
- Desired: experience setting up, maintaining, and tuning infrastructure for AI/ML workloads
- Desired: experience with HPC infrastructure supporting a scientific / research / pharmaceutical / bioinformatics / cheminformatics environment
