Skills and Qualifications
Languages
- Python
Frameworks
- PyTorch
Tools
- Docker
- DVC
Platforms
- MLFlow
Featured Projects
Projects
Video Podcasts
Data Scientist
An application for breaking text into paragraphs for a podcast startup. Kirill undertook the task of creating a system to automatically break text up into semantic sections. Kirill used various networks of the transformer architecture as well as different variations of Bert. He implemented the facility to allow the system to split the podcast’s text into semantic pieces and a function to automatically generate a title.
Hitfactor
Data Scientist
A project that was required to find athletic starting signals or shots on a soundtrack. Kirill helped develop an architecture using convolutional and recurrent neural networks. He was also required to allow the system to be able to separate shots coming from other directions, which are not important.
Vehicle detection and license plate recognition
Data Scientist
A system to solve the problem of finding traffic violations. For the solution, it was necessary to detect cars, compare and identify cars from different camera angles, find violations, and recognize license plates. The project required different OCR approaches for license plate recognition, etc.
