Senior Data Scientist

This is a fully remote position, open to applicants in Germany, +1 more country.

📋 Description

• Develop and enhance censored bid-landscape models for estimating clearing-price distributions from partially observed auction data.

• Create real-time win probability estimation models that respond to bid pricing dynamics.

• Design and execute hierarchical lift estimation models utilizing confidence-bound selection strategies.

• Construct conversion propensity models leveraging sparse, delayed, and aggregate-only labels.

• Formulate look-alike audience modeling techniques employing positive-unlabeled learning and embedding-based nearest-neighbor methods.

• Implement advertiser-level calibration strategies while monitoring ranking and calibration quality independently.

• Develop robust offline evaluation frameworks using inverse-propensity scoring, doubly-robust estimators, and importance reweighting.

• Establish exploration strategies and propensity logging methods to guarantee reliable downstream correction and evaluation.

• Create constrained optimization mechanisms tailored for campaign objectives, pricing constraints, and volume targeting.

• Contribute to data diagnostics, capability assessments, and evidence-driven model recommendations.

• Collaborate with the Customer team during post-launch tuning and performance validation processes.

• Prepare technical documentation and knowledge transfer materials for the Customer’s internal data science team.

• Engage in architecture discussions and contribute to scalable ML platform design decisions.

⛳️ Requirements

• Over 5 years of experience in Machine Learning or Data Science with production-grade models evaluated against business KPIs.

• Proficient in Python, including libraries such as numpy, pandas, and scikit-learn.

• Strong SQL skills and experience with large-scale datasets.

• Extensive practical experience with XGBoost, LightGBM, or CatBoost.

• Solid understanding of regularization, calibration methods, and handling categorical features.

• Comprehensive knowledge of probability, statistics, confidence intervals, and statistical power analysis.

• Experience in feature engineering for structured and behavioral datasets.

• Practical experience with Spark or PySpark.

• Familiarity with experimentation frameworks and A/B testing methodologies.

• Experience with temporal splits, leakage detection, drift analysis, and slice-based metrics.

• Understanding of explainability techniques, including SHAP and permutation importance.

• Upper-Intermediate English proficiency or higher.

• Experience in AdTech modeling, including CTR/CVR prediction, bid-landscape modeling, audience segmentation, and RTB mechanics is an advantage.

• Background in working with sparse, delayed, or censored labels is beneficial.

• Knowledge of attribution modeling, survival analysis, and positive-unlabeled learning is a plus.

• Practical experience with counterfactual and off-policy evaluation techniques is a bonus.

• Familiarity with isotonic regression and Platt scaling is advantageous.

• Experience with hierarchical, empirical-Bayes, or partial-pooling models is a plus.

• Understanding of constrained or multi-objective optimization approaches is preferable.

• Experience with uplift modeling and causal inference methods is a plus.

• Familiarity with Vertex AI or similar managed ML training environments is a plus.

• Publications, competitive modeling achievements, or open-source contributions related to Machine Learning or AdTech are a plus.

• Strong analytical and problem-solving capabilities.

• Ability to thrive in a data-driven environment.

• Excellent communication and stakeholder management skills.

• Capability to explain complex modeling decisions to both technical and non-technical audiences.

• Proactive mindset with a strong ownership mentality.

• Attention to detail and scientific rigor in experimentation and evaluation.

🏝️ Benefits

• Availability of remote work.

• Opportunity to engage with technically challenging products.

• Collaboration with seasoned engineers and data scientists.

• Direct impact on large-scale production systems.

• Knowledge transfer and collaboration opportunities with the Customer’s internal data science team.