TL;DR
HPC Architect: Architecting and deploying scalable HPC clusters using AWS ParallelCluster and Amazon EC2, ensuring seamless integration with on-premises schedulers with an accent on high-performance networking solutions, including AWS Direct Connect and Elastic Fabric Adapter (EFA), to support low-latency MPI applications. Focus on high I/O workloads (Lustre/FSx patterns), observability, reproducibility and operational excellence.
Location: Remote
Company
Quantori is an international team.
What you will do
- Implement high-performance networking solutions, including AWS Direct Connect and Elastic Fabric Adapter (EFA), to support low-latency MPI applications
- Architect and deploy scalable HPC clusters using AWS ParallelCluster and Amazon EC2, ensuring seamless integration with on-premises schedulers.
- Define and maintain containerization standards using Apptainer (Singularity) or related technologies to ensure binary compatibility across heterogeneous hardware and environments
- Design high I/O workloads (Lustre/FSx patterns), observability, reproducibility and operational excellence
Requirements
- Knowledge of HPC-with-cloud integration strategies, experience with technical feasibility planning
- Strong understanding of regulated enterprise R&D environments
- Understanding of scientific computing reproducibility requirements, relying on enterprise standard tooling
- Understanding of vendor constraints (e.g., licensing restrictions; feasibility of ephemeral Lustre with locked commercial applications; execution of commercial applications over cloud resources)
- Deep understanding of Amazon VPC networking, peering and security group configurations
Nice to have
- Experience in life-sciences
- Experience in specific applications like Schrodinger, Jupyter, Matlab
- Experience with workflow engines
Culture & Benefits
- Remote or office work
- Flexible working hours
- Healthcare benefits: medical insurance and paid sick leave
- Continuous education, mentoring, and professional development programs
- A team with an excellent tech expertise
- Certifications paid by the company
