We invite researchers in the field of machine learning to the Recommender Systems and Personalization team of the Center for Practical AI. The Center for Practical Artificial Intelligence is a strategic division of Sber, creating science-intensive technologies.
We create new algorithms for recommender systems and adapt SOTA approaches for the tasks of the bank and partner companies, helping businesses implement new technologies. We also develop our own open-source frameworks and publish the results of our research at international A/A* level conferences. We invite you to join research and applied projects in the field of recommendation technologies, DL models for sequential data, and approaches at the intersection of LLM and RecSys.
The first stage of selection for this vacancy is an interview with an AI recruiter. After applying, expect a message from them on Telegram; the conversation will take about 10 minutes. The AI recruiter's task is to clarify missing details and speed up the review of your candidacy. The AI recruiter is just starting their journey, so we ask for your understanding. Your experience and participation will help make it convenient and useful for everyone!
Our tasks:
- developing the RePlay library for recommender systems
- developing and improving recommender system algorithms
- developing and training transformer models on event sequences
- conducting research in the field of sequential recommendations, LLM for recommendations, foundation models for event data
- writing articles and speaking at leading global and Russian conferences.
We expect:
- 2+ years of experience in a field related to RecSys / DL / NLP
- ability to write readable and well-structured code in Python
- ability to work with Git
- knowledge of standard libraries and frameworks (numpy, pandas, polars, scikit-learn, pytorch, pytorch-lightning, transformers)
- skills in organizing and reproducibly conducting ML experiments (Hydra, Clearml, Mlflow, W&B, Neptune)
- experience in developing and training models using PyTorch
- knowledge of mathematics, deep understanding of classical ML and modern DL methods, broad ML knowledge in general
- understanding of modern Transformer-based architectures for working with sequences
- knowledge of key recommendation approaches and methods for their evaluation
- ability to read, analyze, and reproduce scientific articles.
Will be a plus:
- experience in successful participation in ML competitions and hackathons
- experience in writing scientific articles, publications
- understanding of the principles of modern LLMs and LLM agents.
We offer:
- a comfortable modern office in Moscow, near Kutuzovskaya metro station
- hybrid work format
- annual salary review and annual bonus
- corporate gym and rest areas
- over 400 educational programs from SberUniversity for professional and career development
- voluntary medical insurance, preferential insurance for family, and corporate pension program
- a flexible discount on mortgage loans, equal to 1/3 of the Central Bank's key rate
- a free SberPrime+ subscription, discounts on products from partner companies.
