GPU-HPC Engineer for Task-Based Interferometric Pipelines

HiPEAC; Bordeaux; Sur place; EUR 55 000 - 75 000; Plein temps

Résumé du poste

Join Inria Bordeaux as an Engineer F/M to adapt scientific codes to a task-based runtime system in the DDF-pipeline, boosting high-performance calibration and imaging for interferometric astronomy.

You will implement GPU offloading, manage distributed and multicore parallelism, and collaborate with the PLER.RION team and LAB partners to ensure scientific correctness across configurations.

Qualifications

• Strong proficiency in C and Python for scientific computing.
• Experience with Linux development, Git workflows and CI processes.
• Fluency in parallel, distributed and GPU programming on HPC platforms.
• Knowledge of code validation and verification.

Responsabilités

• Adapt calibration and imaging pipeline components to the target task-based runtime system.
• Delineate tasks and data structures to maximize scheduling opportunities and minimize overhead.
• Adapt data structures to maximize asynchronous data transfers and minimize dependencies.
• Preserve and verify scientific results after transforming the application across configurations.
• Identify high-potential tasks for GPU acceleration and implement corresponding routines.

Connaissances

C programming

Python programming

Linux development

Git & CI

Parallel programming

GPU programming

Code validation

Outils

Git

Description du poste

Join Inria Bordeaux as an Engineer F/M to adapt scientific codes to a task-based runtime system in the DDF-pipeline, boosting high-performance calibration and imaging for interferometric astronomy.

You will implement GPU offloading, manage distributed and multicore parallelism, and collaborate with the PLER.RION team and LAB partners to ensure scientific correctness across configurations.