RTB House owes its market position to an advanced, innovative bidding system based on Deep Learning. As an ML Researcher , you will be responsible for its development.
Your tasks:
- Designing and implementing models, most often of deep neural networks, used to predict the behavior and preferences of Internet users
- Analysis of the latest works in the field of Machine Learning
- Conducting and analyzing A/B tests of new solutions
- Developing and testing new approaches to modeling key issues, such as bidding in first-price auctions
Technologies used:
- Python, Java, Scala
- PyTorch, NumPy, Pandas
- Jupyter Notebooks
Our expectations:
- 4+ years of hands-on experience with Machine Learning / Data Science
- Interest and willingness to constantly develop in the area of Machine Learning
- Knowledge of statistics and probability
- Proficiency in programming
Additional advantages will be:
- Experience in the ML industry
- Experience in designing recommendation systems
- Practical knowledge of the tools we use
Sample topics:
- Implementation and analysis of new model architectures, such as convolutional networks, transformer
- Analyze the difference in the distribution of training and production data. Estimating the scale of the problem, identifying the best solution, analysis, implementation, deployment
- Experiments with the use of transfer learning
- A completely new approach to generating product recommendations that is free from proxy targets and uses contrastive loss function
In return, we offer:
- A very attractive salary
- Work in a team of enthusiasts with experience in machine learning who willingly share their knowledge and skills
- Extremely flexible cooperation conditions – you work when it is convenient for you and devote as much time as you can; you can work fully remotely
- Access to the latest technologies and the possibility of real use of them in a large-scale and highly dynamic project;
- Possibility of using the possessed knowledge and competences in practical applications – while optimizing algorithms supporting hundreds of millions of Internet users and monthly buyers of billions of advertising views in the RTB model, based on extensive data sets
- The effects of your work immediately visible in the company’s business results
- Possibility to publish your results
probability Scala statistics pandas Data Science Python transformer deep-learning NumPy Java transfer-learning Software Developer PyTorch