Senior Data Scientist-Data & Personalization Platform (Hybrid)

What you’ll bring to the table: (Responsibilities)

• Design and own the scoring framework that ranks eligible offers per member — defining features, weighting logic, and validating against business outcomes, then evolving it from deterministic scoring toward ML-driven personalization.
• Lead segmentation and feature pipelines: member group construction, derived attributes, bucketing strategy, and reusable feature sets for eligibility evaluation and targeting.
• Architect and optimize large-scale batch processing workflows handling hundreds of millions to billions of records, including partitioning, bulk ingestion, and performance tuning.
• Define and operate SLAs across the pipeline: batch completion, feed delivery, attribute freshness, and assignment turnaround.
• Provide architectural guidance on a near-real-time assignment API layer and its integration with the broader batch pipeline.
• Define and maintain data contracts with downstream consumers (analytics marts, dashboards, adjacent platforms) and oversee the incremental build-out of analytics data models.
• Translate between business stakeholders (product, marketing, finance) and the engineering team — comfortable holding a business conversation and a technical one in the same meeting.
• Document architecture, data models, pipeline logic, and feature generation processes to reduce key-person dependency and support team continuity.
• Shape the future roadmap for personalization and recommendations, including A/B testing frameworks, behavioral modeling from member activity, and the role of ML and generative AI in assignment and eligibility.

Do you have the right mix of ingredients: (Requirements)

• Master’s degree in data science or related field
• 5+ years of experience in a data science role
• Strong technical foundation across both data science and data engineering — this role owns and directs production pipelines, not just analysis.
• Proven experience designing and leading large-scale data processing systems (hundreds of millions to billions of records), including batch architecture, partitioning, staging, and performance optimization.
• Track record designing activity-based segmentation and tiering frameworks (e.g., RFM-style models, engagement tiers, merchant activity classifications) — from threshold definition through refresh cadence and validation against business outcomes.
• Hands-on background building scoring, ranking, or recommendation frameworks, with feature selection, weighting strategies (rule-based, heuristic, or ML-driven), and evaluation against business objectives; experience evolving such systems from deterministic scoring toward ML-based personalization.
• Experience designing and managing customer segmentation pipelines and feature generation at scale, including the lifecycle management of member groups, derived attributes, and reusable feature sets.
• Experience with workflow orchestration (Airflow/MWAA or equivalent) and AWS data services (S3, Glue, Aurora/PostgreSQL).
• Strong SQL and Python skills — able to review, guide, and produce production-quality data pipeline code.
• Understanding of event-driven architectures and Kafka-based data replication patterns.
• Experience with or strong understanding of real-time or near-real-time data systems, and the ability to provide architectural guidance even when not the primary builder.
• Ability to define pipeline SLAs and data freshness guarantees, including monitoring, alerting, and incident response for batch and near-real-time workflows.
• Experience working with large-scale member or customer data in a personalization, targeting, loyalty, or recommendation context.
• Demonstrated ability to work cross-functionally and influence without authority; self-directed and able to own a workstream end-to-end with minimal oversight.
• Strong written and verbal communication; able to produce clear documentation and present findings to non-technical audiences.

Nice to Have:

• Experience with offer, loyalty, dining, or hospitality platforms.
• Familiarity with Scala or JVM-based systems, particularly in real-time API or microservice contexts that integrate with data pipelines.
• Experience with analytics engineering (dbt or similar) or oversight of BI data model layers.
• Familiarity with CDC-based replication patterns and data synchronization between systems.
• Familiarity with ML model deployment and serving (AWS SageMaker, Bedrock, or equivalent), A/B testing frameworks, and an informed point of view on how foundation models and RAG-based architectures can be applied to personalization and recommendation at scale.

Tech Stack:

• Aurora (PostgreSQL) / SQL
• Python
• AWS (S3, Glue, Redshift, CloudWatch, MSK)
• Airflow (MWAA)
• Kafka / Kafka Connect
• Scala / Kubernetes — architectural awareness required; not expected to be the primary builder