Senior GPU Software Engineer in ML at Intel Corporation (Gdask, Poland)
Location: Gdask, Poland
Type: Full Time
Created: 2021-07-31 05:01:01
Come join a team of GPU software engineers dedicated to making Intel Graphics successful in data centers. We write optimized GPU kernels for neural networks for pre-silicon architectures. Our purpose is to confirm viability of the future hardware platforms and to give other teams early access to highly-optimized code.Your responsibilities will include all stages and areas of software development:
- Deep learning topology performance research and performance modeling
- Source code development, review and maintenance in C++ and Python. Both for pre-silicon and post-silicon hardware.
- Machine learning workloads performance analysis and hardware features validation
- Performance optimization of GPU software components
- Mentoring and overseeing the work of junior colleagues
- Leading a small team or playing a key role in planning and implementation of workloads
- Competitive pay, bonuses and benefits
- Flexible working hours and hybrid home-office working schedule
- Relocation support
- Opportunities to transform technology and create a better future, by delivering products that touch the lives of every person on earth
- Collaborative environment allowing the brightest minds in the world to come together to achieve exceptional results with great team atmosphere
- Opportunities for professional development and growth - including tech trainings, learning platforms and possibility to shape future hardware architectures
You should consider applying if you:
- Have parallel programming experience
- Understand deep learning primitives and basics of linear algebra
- Enjoy problems that can't be solved by searching StackOverflow
- Work well in a team and across teams
You will enjoy this position even more if you have:
- System-level language programming experience: C, C++, others
- GPU programming experience: DPC++/OpenCL/CUDA/ROCm/C for Metal, etc.
- SIMD programming experience: SSE/AVX/Neon, etc.
- Assembly reading or writing experience: x86-64/ARM, etc.
- Experience with a deep learning framework: TensorFlow/Caffe/PyTorch/MXNet, etc.
- Experience profiling, benchmarking and optimizing implementation of a predefined problem