Senior R & D Engineer
з/п не указана
Вакансия в архиве
Работодатель, вероятно, уже нашел нужного кандидата и больше не принимает отклики на эту вакансию
Machine Learning and CNN Programming Tools Engineering Expert for Computer Vision
We are a leading-edge R&D team developing an advanced Embedded Vision Processor (EVP) driving computer vision, image analytics and augmented reality applications for embedded systems. The EVP combines general-purpose vision vector processing and a specialized programmable engine supporting deep learning approaches, leveraging convolutional neural networks (CNN).
We are looking for an experienced engineer to work on the development of programming and analysis tools for the programmable CNN engine. The programming environment supports deep learning frameworks like Caffe, Tensorflow and others.
In this position, you will have responsibility for the following activities:
- Development of optimized CNN engine mapping tools and neural network compiler
- Contribution to the development of performance analysis and resource utilization tools for the CNN engine
- Mapping and benchmarking of leading-edge CNN graphs on the CNN engine
- Close interaction with System, Hardware and Software architects to best leverage the features of the CNN engine and runtime
- Bachelor´s or Master’s degree in Computer Engineering-related field with relevant experience with the development of embedded software tools.
- Experience with the use and development of programming tools for embedded systems (compilers, debuggers, ISS, profiling tools etc.)
- Compiler development experience is highly preferred
- Knowledge of operating system fundamentals
- Strong programming skills in C and C++
- Experience with assembler language and Python
- Good teamwork and communication skills
- Good understanding of vision, imaging or video applications, and the use of embedded processors (GP-GPU, DSP, FPGA, multi-core platforms)
- Knowledge and experience with the use of highly parallel processors, GP-GPUs
- Basic knowledge of machine learning principles and frameworks