Deep Learning Engineer
Python
Deep Learning Engineer
Warszawa
Type of work
Undetermined
Experience
Mid
Employment Type
B2B, Permanent
Operating mode
Office
Tech stack
PyTorch
Python
NumPy
Deep Learning
Computer Vision
Docker
Pandas
Machine Learning
AWS
Job description
Online interview
Company description
ReSpo.Vision revolutionizes football with the help of AI. We employ bleeding-edge Machine Learning and Deep Learning models to extract extremely detailed players’ positional information. We then crunch this data and build predictive models helping clubs, bookmakers & media in player/team analytics, live tactical recommendations and game events’ probability assessment.
We have Headquarters in Warsaw, Poland but are currently running a semi-remote work setup - so we are flexible regarding the location, although we do have preference for a Warsaw based-person.
Our team consists of highly skilled Data Scientists, among them 3 Kaggle masters with multiple awards under their belt and strong academic experience. Tough problems & challenges guaranteed!
About
In this role you will be working on application of state of the art Computer Vision Deep Learning algorithms for football data based on video streams.
You will be responsible for developing novel CNN models and tuning existing ones, optimizing their accuracy, performance and creating data processing pipelines. You will have to be able to efficiently test different hypotheses and iterate over ideas quickly.
We are open to ideas and give freedom in the taken approach and experimentation and expect that you will be able to work on the given tasks and deliver results with little supervision. Full-time position. Work mostly from the office. We are flexible in terms of contract type (B2B & UoP/ UZ possible).
Responsibilities/Tasks
- Tuning & optimization of existing CNN models
- Development of novel task-specific detection & segmentation CNN architectures
- Building data pipelines for DL models
Requirements
- Proficiency with numpy, scipy, pandas
- Proficiency with Pytorch
- Experience working with image/video data
- Experience in building pipelines for CNN models
- Experience with object detection & image segmentation using DL
- Good command of English
- Running independent projects end to end
- Self-directed research
- B2B or UoP/UZ contract type
Preferred Qualifications
- Practical knowledge of Linux
- Practical experience using SOTA CNN models & libraries
Tech Stack
- Python
- Pytorch
- AWS
- Docker
Benefits
- Dynamic start-up environment with interesting scientific challenges & bleeding edge technology and models (this time for real)
- Opportunity to build real-world deployable machine and deep learning-based products changing the world of Football
- International projects & collaborations with top Football Clubs & Federations
- Ability to directly influence development direction of a product
- Team Kaggle competitions (we have 3 masters) & hackathons
- Office in the strict center of Warsaw
Check similar offers
