HPC Engineer – GPU-Accelerated Astronomy Pipelines (Remote)

Inria, the French national research institute for the digital sciences; France; Hybrid; EUR 30,000 - 34,000; Full time

Benefits offered by this job

Public transport reimbursement

Extended leave (7 weeks + RTT days)

Teleworking option

Professional training

Job summary

Inria, the French national research institute for the digital sciences, is seeking an Engineer in computer science to advance the DDF-pipeline with a task-based runtime on GPU-based HPC platforms as part of project PLER.RION.

You will collaborate with the Bordeaux Astrophysics Laboratory, Paris Observatory, and the ECLAT Joint Laboratory to push high-performance interferometric radio astronomy processing forward.

Qualifications

• Master degree in HPC or astrophysics with strong CS background.
• Strong interest in optimization, performance, and rigorous scientific processes.
• Willingness to work in a collaborative research environment.

Responsibilities

• Adapt calibration and imaging pipeline to the task-based runtime system.
• Delineate tasks and optimize granularity for scheduling.
• Adapt data structures for asynchronous transfers.
• Preserve correctness across configurations and transitions.
• Identify GPU-offload opportunities and implement them.

Skills

C programming

Python programming

GPU programming

Parallel programming

Linux development

Git

CI

English proficiency

Education

Master's degree in high performance scientific computing

Master's degree in Astrophysics with CS experience

Tools

killMS

DDFacet

CUDA

MPI

Job description

Inria, the French national research institute for the digital sciences, is seeking an Engineer in computer science to advance the DDF-pipeline with a task-based runtime on GPU-based HPC platforms as part of project PLER.RION.

You will collaborate with the Bordeaux Astrophysics Laboratory, Paris Observatory, and the ECLAT Joint Laboratory to push high-performance interferometric radio astronomy processing forward.