Senior Data Scientist (Product Analytics) — Session Analyzer

Role summary We are looking for a Senior Data Scientist to build the core intelligence layer of our Session Analyzer product: scoring, audience segmentation, and prescriptive recommendations for pharmaceutical marketing teams. You will operate as a “full-stack” data scientist across data validation, data engineering (prod pipelines), ML modeling, causal measurement, and AI-assisted workflows (agents for insights and data diagnostics). You will work directly with the Product Lead and Tech Lead.

What you will do Design and ship scoring, segmentation, and recommendation features based on pixel session/event data, NPI-enriched profiles, and 3rd-party datasets. Build a data quality & validation framework from scratch: tests, monitors, anomaly detection, drift tracking, and root-cause workflows across the data chain. Deliver batch PoC (days latency) and evolve to near-real-time MVP (hours latency) pipelines and feature tables. Develop and evaluate ML models for engagement prediction and audience targeting; ensure calibration, stability, and business-relevant lift. Define and implement measurement/attribution approaches to support ROI uplift and explain “what worked / why”. Build AI agents for (1) automated data issue diagnostics and (2) insight/presentation generation, with guardrails and traceability. Produce outputs as score tables, top insights, aggregates/features, dashboards/reports, and APIs for audience activation and CDP/CRM integrations.

Success metrics Improved Engagement score Measurable uplift in campaign ROI Reduced time-to-insight Increased share of HCPs with actionable recommendations Improved coverage/quality signals (e.g., NPI resolve rate progress, event capture reliability)

Requirements (must-have) 5+ years in Data Science / Analytics roles with production ownership. Strong Python + SQL; hands-on building production pipelines and data products. Experience with event/session analytics, aggregation modeling, and working with noisy web/pixel data. Proven ability to build propensity/engagement models, segmentation, and recommendation logic; strong evaluation practices. Working knowledge of causal inference / attribution methods for marketing effectiveness. Experience with GCP / BigQuery in production; familiarity with ClickHouse/Postgres is a plus. Ability to operate end-to-end: problem framing → data validation → modeling → deployment → monitoring → stakeholder comms. Hands-on experience building and operating data-driven marketing/advertising analytics products, including event-level measurement, audience segmentation, activation workflows, and performance attribution (AdTech/MarTech environments).

Nice-to-have Experience with agents (LLM-based workflows), automated insights, and safe deployment patterns. Familiarity with data quality tooling (e.g., Great Expectations/dbt tests) and observability.

Soft skills & values (required) Openness Direct feedback, acknowledges mistakes, listens and acts on critique. Internal Locus Takes responsibility for outcomes, analyzes own decisions first, learns quickly from failures. Get it done Delivers end-to-end, removes blockers independently, completes work to production-grade finish. Self-improvement Continuously adopts new tools (incl. AI), improves workflows, increases productivity without heavy training overhead. Ownership Treats the product and system results as their responsibility, not just personal tasks/KPIs; minimizes need for micromanagement.