In this role you will be working on application of state of the art Computer Vision Deep Learning algorithms for football data (from visual & sensor feed). You will be responsible for creating machine learning pipelines, tuning models and choosing data processing techniques for different parts of the system.
The engagement will last for a minimum of 6-8 weeks and will involve weekly workload of 30+ hrs. Working remotely is preferred. The team will be managed by Kaggle master with extensive knowledge in the field and supported from the business side by domain matter experts with extensive executive background.
Required skills:
- Good knowledge of Python
- Good knowledge of numpy, scipy, pandas
- Practical knowledge of Linux OS
- Proficiency with deep learning framework, preferably Pytorch
- Experience in working with image/video data
- Experience in building pipelines for deep learning models
- Experience with Deep Learning techniques (CNN)
- Classification
- Object Detection
- Instance/Semantic Segmentation
- Being up to date with current SOTA (papers & implementations) and preferable practical experience with working with those implementations
- Ability to quickly iterate and adjust to changes
- Ability to effectively design software architecture
- Good command of English
- Fluent Polish preferred
Additional details:
- 100% remote position
- Flexible working hours
- Preferable full time, at least 3/4 FTE