About
Delivery Hero operates a local delivery platform in around 65 countries, providing customers with fast, convenient delivery. Its search functionality spans more than 60 countries and 35 languages, facilitating over 80 million searches daily.
Responsibilities
•Design, build, and productionize deep neural ranking models, owning their full lifecycle from offline experimentation and evaluation to monitoring and iterative improvement
•Use LLM coding agents to autonomously iterate on feature engineering and Learning to Rank architectures; design overnight experiment pipelines and review their outputs
•Innovate in feature engineering and model architecture by applying ranking signals, feature stores, and embedding-based retrieval, and validate new architectures based on the latest research
•Monitor production model health, including model drift, latency degradation, and feature pipeline integrity, and respond to deviations
•Collaborate with Backend Engineers, Data Engineers, and Product Managers on end-to-end improvements; explain ML trade-offs to non-technical stakeholders and contribute to the team roadmap
•Mentor junior and mid-level data scientists, promote experimentation rigor and code quality, and contribute to learning about agentic development workflows
Requirements
•Master's degree in Computer Science, Mathematics, Physics, or a related quantitative field, or a Bachelor's degree with 6+ years of work experience
•4+ years of industry experience as a Data Scientist or Machine Learning Engineer applying ML in high-traffic production environments
•Production experience implementing deep learning ranking architectures, including DCN-V2, MMoE, and Two-Tower; knowledge of multi-task learning, cross-feature interactions, and embedding optimization at scale
•Knowledge of Learning to Rank methods, offline evaluation metrics such as NDCG and MRR, and the challenges of translating offline metrics into online A/B outcomes
•Proficiency with LLM coding agents and agentic CLI tools for autonomous experiment generation, architecture prototyping, and rapid iteration; familiarity with the auto-research paradigm and reviewing large volumes of experiment results
•Expertise in Python and its ML ecosystem, including PyTorch, TensorFlow, and scikit-learn
•Experience with SQL, dbt, GCP or AWS, and large-scale data processing with PySpark or Scala
•Knowledge of the ML lifecycle, including feature stores, vector databases, model versioning, A/B experimentation frameworks, and deployment to high-throughput, low-latency serving infrastructure
•Hands-on MLOps experience with CI/CD for ML pipelines, Metaflow, experiment tracking, model registries, and production monitoring for drift, latency, and data quality
•Ability to navigate ambiguity, translate business objectives into scoped data science projects, and take accountability for outcomes
•Ability to collaborate in globally distributed, cross-functional teams and communicate model behavior, experiment results, and trade-offs to technical and non-technical audiences
Conditions
•27 days of annual leave, plus one extra day in the second and third years of service
•€1,000 educational budget, language courses, parental support, and access to Udemy Business
•Health checkups, meditation, and gym access
•Employee Share Purchase Plan, Sabbatical Bank, public transportation ticket discount, life and accident insurance, and corporate pension plan
•Digital meal vouchers and food vouchers
•Relocation information and support resources for moving to Berlin are available
•Severely disabled applicants with equal qualifications will be given preferential consideration
