GPU Engineer

This is a fully remote position, open to applicants in France.

📋 Description

• Design and enhance GPU kernels utilizing C++, CUDA, HIP, and low-level instructions.

• Explore GPU architecture, memory access, synchronization, and execution bottlenecks.

• Profile and benchmark workloads to detect and confirm performance enhancements.

• Optimize inference operations and facilitate communication across various GPUs.

• Implement and refine new AI models on both NVIDIA and AMD GPUs.

• Assist in creating AI agents for GPU engineering, which will enable automated code optimization, correctness testing, benchmarking, and iterative improvements.

• Assume responsibility for technical decisions and their execution.

⛳️ Requirements

• Significant hands-on experience in developing and optimizing GPU kernels, preferably with CUDA C++ or HIP/ROCm.

• In-depth knowledge of GPU architecture, memory hierarchy, parallel execution, and performance bottlenecks.

• Strong C++ proficiency along with experience in profiling and benchmarking GPU workloads.

• Proven track record of taking ownership of complex technical projects, from implementation to measurable outcomes.

• Enthusiasm for developing AI agents and tools aimed at expediting GPU engineering.

• Experience with PTX, CDNA ISA, multi-GPU systems, low-latency inference, or GPU performance portability is especially advantageous.

• A PhD in HPC, computer science, or a related field is beneficial, especially when supported by substantial engineering experience.

• Candidates should already reside in Paris or other areas within Île-de-France.

• Must be available to join the team promptly.

🏝️ Benefits

• Comprehensive health and wellness programs.

• Opportunities for professional development and growth.

• Collaborative and innovative work environment.