RocketHuntSmart job search
Reach out directly about this role
We are looking for an ML developer who will improve the quality of Alice AI LLM pre-training. This is the area responsible for the core intelligence of the base model and data quality.
The LLM race is accelerating: models are solving problems in mathematics, code, agents, and medicine that only recently required human involvement. At the same time, major improvements in LLMs occur when pre-training is updated and scaled: it is at this stage that the model's fundamental capabilities are laid, upon which quality depends at all subsequent stages.
We believe that the winner of this race is the one who improves pre-trains more effectively and faster, and we are investing in this area: we research ways to improve model quality, conduct computationally intensive experiments, and process and synthesize massive amounts of data.
The result of our work is the Alice AI family of models, available through the service of the same name. By making our model smarter, you will improve the leading and most widespread AI assistant in Russia, which is already used by over 24 million people weekly.
What we have already done: Our current pre-train models have already shown success in previous releases of Alice. For example, we wrote on Habr that we demonstrate superior quality on factual and cultural datasets, measurements, and more. We discussed the research that formed the basis of these improvements at Practical ML Conf 2025.
What will you be doing? Model reasoning capabilities have proven benefits in areas such as mathematics, code, and agents, but leading players are achieving significant quality gains through reasoning on a broader range of tasks, from finance to computer games. Our goal is to make Alice's reasoning model the strongest in Russia and beyond.
Experiments with Pre-train Models Conduct and analyze computationally intensive experiments with the pre-train stage. Test hypotheses about the sources of reasoning capability growth within various aspects of intelligence. Defining and selecting the experiment setup is a creative stage.
Data Pipelines Build efficient pipelines for collecting, synthesizing, processing, and filtering training data in new domains. The built pipelines will need to be scaled and made more efficient.
Modifying the Training Process Modify the model training process to improve reasoning tasks. Implement changes in the training pipeline and evaluate their impact on quality.
Expert Models Train verifier models to assess the quality of data and responses. Train models to generate training data that is SOTA in its narrow domain. Train filter models to discard low-quality data.
Scaling Research Investigate the scaling laws of model quality in terms of data and computation. Determine how scaling affects the growth of reasoning abilities across different domains.
More about ML at Yandex — on the Yandex for ML channel
Яндекс
Russia14 days ago
Employment
Full-time
AI writes it from your resume. You only have to send it.
By company and country
