Senior GPU/CUDA Performance Engineer

TMC; Eindhoven; On-site; EUR 90,000 - 130,000; Full time

Benefits offered by this job

Profit sharing (Employeneur) lower

Personal training budget

Technical coaching

Strong network of high-tech pros

Work on complex, innovative projects

Freedom to shape your development

Job summary

TMC is seeking a Senior GPU/CUDA Software Engineer to design and optimize high-performance C++/CUDA software for GPU-accelerated applications. You will work on signal/image processing, numerical modelling, and AI workloads, translating mathematical algorithms into efficient GPU implementations.

The role involves profiling, kernel optimization, memory management, and CPU-GPU data transfers, with close collaboration across FPGA, electronics, firmware, systems, algorithms, and AI teams to shape

Qualifications

• Master’s degree or PhD in a technical field such as CS, EE, or Math.
• Strong professional experience with C++ and CUDA.
• Deep hands-on CUDA GPU programming and performance optimization knowledge.
• Experience with memory hierarchy, bandwidth, and occupancy.
• Familiarity with profiling and performance-analysis tools.
• Solid understanding of object-oriented software design.

Responsibilities

• Design, develop, and optimize high-performance C++ and CUDA software for GPU-accelerated applications.
• Translate computationally intensive algorithms into efficient GPU implementations.
• Profile and optimize CPU-GPU data transfers and kernel performance.
• Improve latency, throughput, and scalability across heterogeneous systems.
• Collaborate with multidisciplinary teams including FPGA, firmware, AI, and systems engineers.
• Contribute to hardware-software partitioning and system architectures.

Skills

CUDA

C++

GPU programming

Performance optimization

Profiling tools

Parallelization

Memory optimization

CPU-GPU transfer

Education

Master's degree

PhD

Tools

Nsight

CUDA Toolkit

Job description

TMC is seeking a Senior GPU/CUDA Software Engineer to design and optimize high-performance C++/CUDA software for GPU-accelerated applications. You will work on signal/image processing, numerical modelling, and AI workloads, translating mathematical algorithms into efficient GPU implementations.

The role involves profiling, kernel optimization, memory management, and CPU-GPU data transfers, with close collaboration across FPGA, electronics, firmware, systems, algorithms, and AI teams to shape