Детали вакансии

Backend Developer for Feature Store at Yandex. Focuses on ML models for ad ranking, addressing training-serving skew. Integrates with Search and Ads services, develops Python interfaces, and optimizes C++ computations. Requires C++, Python, and experience with high-load systems.

Hundreds of ML models are responsible for the quality of ad ranking. The ML development team continuously trains new, more advanced models. One of the complex problems that significantly affects online experiment metrics is the discrepancy between training and serving data (training-serving skew). In Feature Store, we are developing a meta-layer. It allows ML developers to construct and reuse features for models in the context of low latency (real-time application) and high throughput (retraining on a cluster). What tasks await you: • Integration with Search and Ads services Feature Store has become a standard for implementing neural networks in Ads, and you will help us further integrate into various services, including Search and Ads, as well as adapt to the needs of new model classes (generative transformers). You will have to solve various problems: from optimizing C++ computations to developing Python interfaces for users and integrating with internal services (YQL, Model Registry, model services). Your knowledge will help improve user experience, speed up the launch of new models, and save resources.

We expect you to: • Be proficient in C++, familiar with Python • Have worked with high-load systems • Know classic algorithms and data structures

Yandex is a community. There are sports clubs, a book club, and an esports community here. That's not all — the full list is here: https://yandex.ru/jobs/pages/benefits?utm_campaign=ya_nanimaet