ML Developer for the Plus Loyalty Team

Yandex Plus is a unified subscription for Yandex services, which provides access to music, movies, podcasts, books, games, sports, and other content. It is a large ecosystem project where over 35 million subscribers use diverse features across all ecosystem services every day: they listen to 'My Wave' in Music, get cashback in Taxi, Food, Market, and other Yandex services, and watch movies on Kinopoisk.

One of the focus areas is the development of loyalty scenarios. The number of offers for users is growing daily, and it is necessary to build systems for optimal management of this offering, discounts, and Plus points.

There are quite a few loyalty mechanics, each of which needs to be optimized and as effective as possible without exceeding costs. At the same time, all mechanics must work in the same direction without contradicting business goals.

What tasks await you

Formulating hypotheses and conducting experiments You will propose new hypotheses and test them to improve the effectiveness of loyalty mechanics for our subscribers in the Yandex ecosystem using ML models.

Writing production code in Java We work with classical ML methods, wrap our models in Java services, and deploy them to production. You will write reliable and efficient production code in Java.

We expect that you

• Understand the principles of classical ML
• Know traditional algorithms and data structures
• Are well-versed in Python and SQL
• Can clearly demonstrate research results to the business and argue your point of view convincingly
• Are ready to be responsible for production models in runtime

Will be a plus

• Have deployed ML models to production
• Have experience in Java or C++ development
• Are able to solve problems in the field of Causal Inference