Farouk Ghandour

About Farouk Ghandour

I’m a Data Scientist focused on using analytics, experimentation, and applied modeling to drive product, marketing, and revenue decisions.My work sits at the intersection of data analysis, experimentation, and applied machine learning—turning messy, real-world data into clear metrics, insights, and decision frameworks used by product, marketing, and operations teams. I care deeply about ownership: defining success metrics, designing experiments, building models, and translating results into actions that scale.At Thermo Fisher Scientific, I build and deploy analytics and modeling solutions across churn prediction, customer segmentation, forecasting, experimentation, and automated reporting, supporting high-impact decisions with measurable business outcomes. I focus on maintainable systems and repeatable insights rather than one-off analysis.I hold a Master’s in Data Science from UC Berkeley and enjoy working on ambiguous, ownership-heavy problems where data directly shapes strategy. I’m especially interested in roles that combine product thinking, experimentation, and analytical depth.Open to conversations around Data Scientist, Product Analytics, and Decision Science roles.

Experience

1. Data Scientist

   Thermo Fisher Scientific

   Sep 2024 — Present · San Francisco, CA, US

   Designed and operationalized end-to-end applied ML systems for customer intelligence, spanning data ingestion, feature engineering, model training, and production inference.• Built predictive churn and anomaly detection models (Random Forests, decision trees) to identify at-risk customers and unusual purchasing behavior, surfacing ~$21M in potential churn risk.• Developed and integrated LLM-powered pipelines (BART, OpenAI) for sentiment analysis and domain resolution, automating feedback classification and improving downstream data quality.• Designed and analyzed A/B and multivariate experiments to evaluate model-driven strategies, translating results into actionable recommendations that improved engagement and retention.• Implemented RFM-based segmentation and customer journey logic to inform targeting strategies and downstream decision systems.• Automated reporting pipelines and dashboards to surface model outputs and KPIs in near real time, enabling faster decision-making and reducing manual reporting effort by ~20 hours per month.

Education

• Purdue University

  Specialization , Applied Generative AI

• Suffolk University

  Bachelor's degree, Big Data Analytics/ Minor in Finance

  2014 — 2018

• University of California, Berkeley

  Master's degree, Data Science

Skills

• Microsoft Office
• Customer Service
• Microsoft Excel
• Leadership
• Microsoft Word
• Powerpoint
• Research
• Public Speaking
• Social Media
• Marketing