Software Engineer/GPU Programmer KU Leuven

The TRACE (Toyota Research on Automated Cars in Europe) lab of KU Leuven in Belgium is the central node of a European network that carries out research on automated cars for Toyota. Other partners of the network include computer vision groups at the University of Cambridge, the Max Planck Institute at Saarbrücken, and the Technical University of Prague. For TRACE-Leuven we are hiring a Software Engineer/GPU Programmer Responsibilities TRACE-Leuven has an immediate opening for one or more software engineers with specialized on GPU programming. The position provides the opportunity to deploy image processing algorithms on GPUs, for autonomous driving. This will happen in close collaboration with several Toyota teams (Europe, USA, and Japan). The engineer will work among a highly skilled team of scientists to help build libraries and software for real-time environments, operating close to the actual development of production-ready solutions. The candidate should have a software background with multi-threading experience. He or she will be required to port threaded C/C++ signal and image processing algorithms to GPUs in a robust, optimized format for both real-time and backend applications. The candidate must have a well-rounded understanding of common bottlenecks and be well-versed in computational architectures and optimization. A strong candidate will demonstrate skills in design for optimal concurrency and memory access, as well as building software for multiple GPU support. Since the GPU-optimized algorithms are just one piece of a larger system, the ability to integrate with hardware, controls, and displays is also important. The candidate will report to the Leader of the Development Team. Profile A qualified Software Engineer has: Master’s degree in Computer Science or similar field Experience in general software engineering. Proficiency in C/C++, GPU software experience, proficiency in CUDA Basic knowledge of GPU architectures and hardware Proficiency with GPU profiling and optimization techniques Knowledge of open-source libraries for GPU-related tasks General knowledge on scripting languages, Python, Bash, Perl, … Experience with Linux and Windows Exceptional Candidates will have: Experience with streams, events and multi-GPU Demonstrate the ability to rewrite algorithms for improved performance Experience with OpenGL/Vulkan Experience with image processing / computer vision More than 3 years experience in the above fields is a plus Offer KU Leuven can offer a permanent and fulltime contract within an innovative environment. The position will be situated at a separate legal entity, thus allowing us to offer maximally competitive conditions. Contact Paul.Konijn@esat.kuleuven.be If you apply for this position please say you saw it on Computeroxy

The TRACE (Toyota Research on Automated Cars in Europe) lab of KU Leuven in Belgium is the central node of a European network that carries out research on automated cars for Toyota. Other partners of the network include computer vision groups at the University of Cambridge, the Max Planck Institute at Saarbrücken, and the Technical University of Prague. For TRACE-Leuven we are hiring a

Software Engineer/GPU Programmer

Responsibilities

TRACE-Leuven has an immediate opening for one or more software engineers with specialized on GPU programming. The position provides the opportunity to deploy image processing algorithms on GPUs, for autonomous driving. This will happen in close collaboration with several Toyota teams (Europe, USA, and Japan). The engineer will work among a highly skilled team of scientists to help build libraries and software for real-time environments, operating close to the actual development of production-ready solutions.

The candidate should have a software background with multi-threading experience. He or she will be required to port threaded C/C++ signal and image processing algorithms to GPUs in a robust, optimized format for both real-time and backend applications. The candidate must have a well-rounded understanding of common bottlenecks and be well-versed in computational architectures and optimization. A strong candidate will demonstrate skills in design for optimal concurrency and memory access, as well as building software for multiple GPU support. Since the GPU-optimized algorithms are just one piece of a larger system, the ability to integrate with hardware, controls, and displays is also important. The candidate will report to the Leader of the Development Team.

Profile

A qualified Software Engineer has:

Master’s degree in Computer Science or similar field
Experience in general software engineering.
Proficiency in C/C++,
GPU software experience, proficiency in CUDA
Basic knowledge of GPU architectures and hardware
Proficiency with GPU profiling and optimization techniques
Knowledge of open-source libraries for GPU-related tasks
General knowledge on scripting languages, Python, Bash, Perl, …
Experience with Linux and Windows

Exceptional Candidates will have:

Experience with streams, events and multi-GPU
Demonstrate the ability to rewrite algorithms for improved performance
Experience with OpenGL/Vulkan
Experience with image processing / computer vision
More than 3 years experience in the above fields is a plus

Offer

KU Leuven can offer a permanent and fulltime contract within an innovative environment. The position will be situated at a separate legal entity, thus allowing us to offer maximally competitive conditions.

Contact

Paul.Konijn@esat.kuleuven.be

If you apply for this position please say you saw it on Computeroxy