Data Scientist for the Recommendations Team
About the Team
Our team develops the recommendations section — the first thing people see when they open Avito.
We write microservices in Python and Go and work with machine learning to better understand users' interests. Currently, we are focusing on developing a new transactional uber-like product for Avito in the "Side Jobs" vertical. The key task is to build and refine recommendation algorithms from scratch for this dynamic market.
We are looking for a strong DS engineer with experience in recommendations or search, who wants to apply their knowledge in a new, emerging environment and bring an outside perspective.
Examples of future tasks:
- Design and develop recommendation algorithms for the side job market from scratch (matching, liquidity management);
- Improve the quality of the recommendation system (candidate selection, ranking) and measure success through A/B tests;
- Develop and adapt core models that are used as a foundation for other models in the ecosystem;
- Research modern approaches from scientific papers and implement them in production;
- Improve data and model infrastructure to increase the efficiency of experiments;
- Analyze numerous A/B tests to make informed product decisions;
- Work closely with product teams to gain a deep understanding of the domain.
It would be great if you:
- Have experience working with modern transformer models and their application in ranking or matching tasks;
- Have achieved high rankings in machine learning competitions (Kaggle and similar);
- Have worked with Airflow for pipeline orchestration;
- Are ready to immerse yourself in a new subject area and propose non-standard solutions.
Working with us is:
- An opportunity to implement your ideas in a project with a multi-million audience;
- A talented team ready to support your initiatives;
- A transparent bonus and incentive system, a competitive salary — the amount will be discussed during the interview;
- Company-funded learning: participation in Russian and international conferences, specialized courses, and access to the MIF online library, literature necessary for growth and development;
- Comfortable working conditions: health insurance (VHI) from day one, fitness, meal compensation;
- A beautiful and comfortable office just two minutes from Belorusskaya metro station, as well as the possibility to work remotely.
