Senior Data Scientist - Delivery Consumer Discovery

About the role

Bolt is looking for an experienced Data Scientist to own the ranking and recommender systems behind how customers discover what to order on Bolt Food. You will shape what each customer sees first on the home screen and across other personalised surfaces, taking ideas from research through to production. The role works closely with product, engineering, analytics and operations teams.

Key responsibilities

• Improve ranking and recommendations for the home screen and other personalised screens, including learning-to-rank for categories and merchants, personalised recommenders, cold-start and exploration, and screen assembly.
• Build recommendations that help customers discover new restaurants, stores and items.
• Turn business targets and guardrails into ranking objectives while keeping customer relevance in mind (multi-objective and full-surface optimisation).
• Design and run A/B tests to measure the impact of ranking and personalisation changes.
• Take models from research to production with Product, Engineering, Analytics and ML Platform, then monitor them.
• Shape technical direction (candidate generation, features, logging, experimentation tooling) and explore sequential and generative recommenders, deploying them only where they earn their complexity.
• Use AI coding tools such as Codex, Cursor and Claude Code to move faster.

Required profile

• 5+ years in data science and machine learning, ideally building products at a tech company; expected to drive technical direction.
• Hands-on experience with large-scale ranking, recommender or personalisation systems; search and IR experience is a plus.
• Solid grounding in experimentation, statistical inference and offline and online evaluation, with a strong product sense.
• Has built ML systems end to end and can align stakeholders when customer experience trades off against commercial targets.

Required skills

• Strong Python and SQL, plus PyTorch and the core ML stack (Pandas, NumPy, scikit-learn)
• Experience with several of: learning-to-rank, collaborative filtering, two-tower or sequential models, embeddings and vector search, gradient boosting, multi-objective optimisation
• Production stack: Python, SQL, Spark, Docker, AWS/SageMaker, Airflow

What we offer

• A competitive salary and stock options.
• Wellness perks and a 1-month paid sabbatical after 5 years.
• Annual company events and smaller team gatherings.
• On-site work at the Tallinn headquarters next to the operations and engineering teams.