Blend is seeking a Senior Data Scientist to join our Intelligent Media Targeting team, helping global clients make smarter marketing and media investment decisions using transaction data, advanced analytics, and machine learning. This is a hands-on, client-facing role where you will design customer segmentation models, build explainable machine learning solutions, and develop data-driven media allocation strategies that translate complex analytics into actionable business insights.
You will work closely with product managers, marketing strategists, and engineering teams to develop scalable analytical solutions that enable personalized customer engagement and optimized media investments.
What You Will Do
- Design, develop, and deploy customer segmentation models using unsupervised machine learning techniques including K-Means, Gaussian Mixture Models (GMM), and DBSCAN.
- Engineer high-quality customer features from large-scale transaction datasets, including multi-window RFM metrics, spend trajectories, recency decay, and temporal behavioral patterns.
- Evaluate and validate clustering models using statistical techniques such as silhouette score, stability testing, and business interpretability.
- Develop explainable AI solutions using SHAP and other interpretability techniques to translate model outputs into meaningful customer personas and business narratives.
- Calculate and interpret key commercial metrics including Spend Index and Wallet Share while understanding card network coverage limitations and data constraints.
- Build data pipelines and analytical workflows using Python, SQL, and modern data science libraries.
- Develop media budget allocation methodologies using heuristic approaches, response curve modeling, and marketing optimization techniques.
- Apply temporal disaggregation techniques such as Denton-Cholette (or equivalent methods) to distribute aggregated budgets into monthly media plans.
- Perform data preparation and filtering using issuer BIN/ICA mappings and merchant hierarchy logic.
- Collaborate with business stakeholders to understand marketing objectives and translate them into analytical solutions.
- Present analytical findings to technical and non-technical audiences through clear visualizations, documentation, and business recommendations.
- Contribute to reusable analytics frameworks, code quality standards, and best practices across the data science team.
- Mentor junior data scientists and actively contribute to knowledge sharing within the practice.
What You Need
- Proven experience building end-to-end machine learning solutions for customer analytics, marketing analytics, or personalization use cases.
- Strong hands-on expertise in Python with experience writing clean, maintainable, and production-quality code.
- Advanced proficiency with pandas, NumPy, scikit-learn, and SQL for large-scale analytical workloads.
- Strong understanding of applied statistics including probability distributions, regression, hypothesis testing, statistical inference, and model evaluation.
- Demonstrated experience designing customer segmentation solutions using clustering algorithms including K-Means, GMM, and DBSCAN.
- Experience engineering behavioral features from transaction or customer activity data.
- Hands-on experience evaluating clustering quality using statistical validation techniques and business metrics.
- Practical knowledge of explainable AI methodologies, particularly SHAP, for interpreting machine learning models.
- Experience working with large transactional datasets and translating analytical findings into actionable business recommendations.
- Ability to independently manage analytical workstreams from problem definition through delivers.
- Strong written communication skills with experience documenting methodologies, assumptions, and analytical findings.
- Comfortable working directly with business stakeholders in a consulting or client-facing environment.
Technical Skills
Programming & Analytics
- Expert-level Python development for data science and analytics
- pandas, NumPy, scikit-learn
- SQL including joins, CTEs, window functions, aggregations, and query optimization
- Git and version control best practices
Machine Learning
- Customer Segmentation
- K-Means
- Gaussian Mixture Models (GMM)
- DBSCAN
- Cluster validation techniques
- Feature engineering
- Explainable AI (SHAP)
- Model evaluation
Statistics
- Regression analysis
- Hypothesis testing
- Statistical inference
- Probability distributions
- Experimental design
- Model performance evaluation
Marketing & Customer Analytics
- RFM Modeling
- Customer Lifetime Value (CLV)
- Behavioral segmentation
- Spend trajectory analysis
- Wallet Share
- Spend Index
- Media budget allocation
- Marketing optimization
Data Processing
- Transaction data processing
- Time-series feature engineering
- Data quality validation
- Large-scale analytical datasets
Nice To Have
- Experience working with payment card, banking, or financial transaction data.
- Knowledge of issuer BIN/ICA mappings and merchant hierarchy structures.
- Experience calculating Spend Index, Wallet Share, or similar commercial analytics metrics.
- Familiarity with temporal disaggregation methods such as Denton-Cholette.
- Experience integrating LLM APIs (OpenAI, Anthropic, or equivalent) for automated persona generation or narrative creation.
- Experience using UMAP or t-SNE for customer segmentation visualization.
- Hands-on experience with Spark, PySpark, or Dask for distributed data processing.
- Experience in Financial Services, Payments, Retail, Loyalty, or Consumer Analytics.
- Exposure to Marketing Mix Modeling (MMM), media optimization, or customer targeting platforms.
- Experience deploying machine learning models into production environments using MLOps best practices.
- Advanced degree (Master's or PhD) in Data Science, Computer Science, Statistics, Mathematics, Economics, or a related quantitative discipline.
