HPC Performance Engineer – CPU/GPU Kernel Acceleration

European Tech Recruit

Paris

Sur place

EUR 90 000 - 130 000

Plein temps

Il y a 18 heures

Soyez parmi les premiers à postuler

Recevez plus de réponses des employeurs

Envoyez un CV adapté au poste en quelques minutes.

Résumé du poste

European Tech Recruit is seeking an experienced HPC engineer for a hands-on role shaping performance for a next-generation computing platform. You will optimize kernels and manage CPU/GPU workloads in large-scale simulations and scientific computing.

The role blends low-level optimization with architectural influence, working close to hardware while contributing to performance strategy and reusable HPC libraries across multi-GPU and distributed setups.

Qualifications

Degree in a relevant field required.
Experience in HPC or performance-critical software.
Strong knowledge of memory hierarchy and parallel programming.

Responsabilités

Create high-performance numerical components for heterogeneous CPU/GPU environments.
Build and tune computational kernels and numerical operators.
Move workloads efficiently between CPU and GPU architectures.
Improve memory access, data layouts, cache behaviour and host-device transfers.
Optimise GPU kernels using CUDA, HIP, SYCL or comparable technologies.
Investigate occupancy, warp utilisation, memory coalescing and compute efficiency.
Accelerate CPU code through multithreading, SIMD and vectorisation.
Improve sparse and dense linear algebra operations.
Develop acceleration technology for numerical solvers and scientific workloads.
Profile complex applications to identify processor, memory and communication bottlenecks.
Design reusable HPC libraries and software components.
Support multi-GPU and distributed computing where needed.
Collaborate with numerical solver, CAE, AI and software architecture specialists.
Contribute to technical direction, engineering standards and long-term platform decisions.