SoftQuantus

About this role

Build the software that takes a workload to a supercomputer and brings its result back. You work on Softquantus Cloud, the service every CPU, GPU and HPC job passes through, on schedulers, containers and the telemetry a user needs to choose where a job runs.

Focus Areas

• SLURM job submission, accounting and queue telemetry
• Containers for shared HPC systems (Singularity or Apptainer)
• MPI and GPU workloads at supercomputer scale
• Usage records that a billing platform can trust

Responsibilities

• Design and build the adapters that submit, track and collect jobs on HPC systems
• Package workloads so that they run the same way across different machines
• Expose queue and load data that a user can read before choosing a machine
• Report measured consumption from Softquantus Cloud to the Research Platform
• Write down limits, failure modes and what happens when a system is saturated

Requirements

• Solid experience operating or developing for HPC clusters
• Strong Python, plus Go, Rust or C++
• Working knowledge of SLURM or a comparable batch scheduler
• Comfort with containers, networking and Linux at system level

Nice to Have

• Experience on EuroHPC or national supercomputing systems
• Experience with quantum circuit simulators on GPUs
• Open-source contributions to HPC tooling

Skills & Keywords

HPC Supercomputing SLURM MPI GPU Infrastructure Paris Remote

Compute and quantum hardware engineering