We are looking for motivated machine learning developers of all seniority levels to join us and work on exciting and innovative projects. Our teams are collaborating with some of the world leading companies in domains of networking, graphics, automotive and machine learning. We invite you to join us and make a difference.
Some of the things you will be working on:
- Developing tools for quantization of Machine Learning models in order to run them on machine learning processors with integer/fixed-point arithmetic
- Implementing low-level machine learning kernels, such as convolution, pooling, activation functions using a custom machine learning compiler
- Optimization of implemented kernels for a custom hardware architecture
What you bring to the table:
- Academic education in computer technology, mathematics or physics (finished BSc / MSc studies)
- Experience in Machine Learning, with accent on Computer Vision
- Knowledge of Machine Learning model internals, such as implementation of kernels used in Computer Vision models (2D convolution, pooling, activation functions, etc.)
- Good knowledge and experience with computer arithmetic
- Good knowledge of Python and C/C++ programming languages
- Good problem-solving, debugging and analytical skills
- Positive energy, motivation to learn and ability to work as part of a team
- Desire for improvement of current skillset in a fast-paced working environment
- Fluent in spoken and written English
What would be great, but not mandatory:
- Experience with major machine learning libraries (PyTorch, Tensorflow, ONNX, etc.)
- Experience with Computer Vision neural network models such as ResNet, MobileNet, VGG, Inception, etc.
- Firm knowledge in calculus, linear algebra, discrete math, numerical analysis
- Experience in Digital Signal Processing (DSP))
- Experience with fixed-point arithmetic
- Experience with version control tools - Git
If you do not have these skills, you will have an opportunity to master them.
We thank every candidate who applies to this position. Though we would like to get to meet as many candidates as possible, not all the candidates will start our selection process. In case we do not contact you, we encourage you to follow our Careers page and stay in touch for future openings.