Act as the technical leader for search ranking, owning model and architecture design for deep neural ranking models (DCN-V2, MMoE, Two-Tower) from offline experimentation through production.
Lead agentic ML research using LLM coding agents (e.g., Claude Code, Gemini) to run overnight experiment pipelines and convert strong signals into production gains.
Improve ranker quality through feature engineering, model architecture work, embedding-based retrieval, and applying the latest research into production systems.
Set standards for production model health across squads, including drift, latency, and feature pipeline integrity, and promote their adoption.
Collaborate with Product and Engineering leads to define vision, mission, and goals for search ranking squads and present results clearly to stakeholders.
Mentor junior and senior data scientists, review and audit their work, and contribute to hiring as a bar raiser.
Requirements
Master’s degree in Computer Science, Mathematics, Physics, or related quantitative field, or Bachelor’s with 6+ years of experience, plus 4+ years as a Data Scientist or Machine Learning Engineer in high-traffic production environments.
Proven production experience with deep learning ranking architectures (DCN-V2, MMoE, Two-Tower), multi-task learning, cross-feature interactions, embedding optimization, and solid foundation in LTR methods and offline-to-online metric bridging.
Proficiency with LLM coding agents and agentic CLI tools for autonomous experiment generation, architecture prototyping, and rapid iteration; familiar with auto-research paradigm and reviewing large experiment results.
Expert Python skills across ML ecosystem (PyTorch, TensorFlow, scikit-learn), plus working knowledge of SQL (dbt), GCP or AWS, big-data processing (PySpark, Scala), and hands-on MLOps including feature stores, vector databases, model registries, CI/CD, experiment tracking, A/B testing, deployment, and monitoring.
Experience leading data science projects in ambiguous, complex problem spaces with long-term product or architecture impact, and making effective trade-offs.
Experience setting technical standards across multiple teams, building consensus among diverse stakeholders, and communicating model trade-offs to both technical and non-technical audiences.
Perks and benefits
Hybrid working model with 2 days per week on the beautiful Berlin campus for face-to-face collaboration.
27 days holiday plus additional days on 2nd and 3rd year of service.
Support for personal development and career growth including €1,000 educational budget, language courses, parental support, and access to Udemy Business platform.
Health initiatives such as health checkups, meditation, and gym memberships.
Financial benefits including Employee Share Purchase Plan, Sabbatical Bank, public transportation ticket discount, life & accident insurance, and corporate pension plan.
Digital meal and food vouchers to enhance team gatherings and personal wellbeing.
Relocation support to Berlin with dedicated resources to ease the move.