For a new, but long-term project in processing real-time networking data, we are looking for a skilled DNN developer. While being a part of the international team, the work will involve working with a Silicon Valley-based company and a talented team of engineers based in the US and Russia.
You'll work in a small dynamic team to implement your new machine learning techniques across a range of applications. This work consists of:
- implementing state-of-the-art Deep learning research,
- making application-specific customizations, and
- compressing nets to run on computationally limited devices.
- Design, develop and optimize DNNs (CNN, RNN etc.) in collaboration with our other machine learning engineers
- Explore novel applications for DNNs
- Curate and improve the dataset
- Test approaches on embedded platforms that optimize run time and size
A successful candidate must have:
- Strong mathematical background
- Excellent C++, Python skills
- Graduate work focused on signal processing, machine learning, deep learning or related math grad work
- History of projects in object detection/classification using a convolutional neural net
- Experience with parallel or accelerated computing (OpenCL, CUDA)
- Experience writing unit tests and testable code
- Healthy awareness of current ML/deep learning/CVPR research literature
- Fluency with current frameworks (Caffe, Theano, Torch, Tensor Flow)
- Familiarity with traditional Computer Vision methods
Several locations in Russia (Saint Petersburg, Moscow and Samara). We can help with relocation if needed. This is a unique opportunity to work directly with one of the best teams in the Valley.