Gabriel Sanson
Data Scientist
Data Scientist in Atlanta, USA. Currenlty working at Visa Consulting and Analytics team helping Visa merchants clients with actionable insights from VisaNet data.
Get in TouchAbout Me
I'm a Data Scientist who loves digging into data to uncover cool insights. Whether it's for work, personal projects, or just out of curiosity. I enjoy using my skills to solve problems and make sense of the world around me.
Featured Projects
Work Experience
Visa
Data Scientist
May 2025 — Present- Led 10+ end-to-end analytics projects for major merchant clients, analyzing VisaNet transaction data to identify customer trends and deliver clear, actionable recommendations—contributing $1.5M+ in consulting revenue
- Built an optimization model to meet Visa’s compliance rules, to ensure client datasets could be shared safely and without restrictions enabling the onboarding of multiple new clients
- Created in-house models to predict key demographic attributes (such as age, income levels, and spending patterns) directly from VisaNet transaction behaviors, removing the limitations of third-party datasets and allowing insights for both credit and debit customers
RippleWorks
Data Scientist
Jan 2025 — Mar 2025- Designed and deployed an AI pipeline for the U.S. Space & Rocket Center, leveraging dynamic survey classification, sentiment analysis, and LLM-based summarization, securing a $65K annual contract
- Built a scalable Databricks SQL database and custom ETL pipeline, automating data ingestion and transformation for AI-driven insights
- Developed a reusable NLP python package for survey data and text responses, standardizing classification, sentiment analysis, and summarization workflows reducing manual effort
- Secured a major enterprise client with a $100K+ annual contract by integrating Garmin’s webhook API, enhancing Rippleworx’s product with employee sleep and exercise data
BestBuy
Data Science Intern
Jun 2024 — Aug 2024- Led a customer churn analysis model by leveraging customer comments and large language models (LLMs) to identify key drivers of churn, providing actionable insights for retention strategies
- Build Logistics Regression model using LLM extracted features to rank the importance of customer issues
Bradesco Bank
Data Analyst
Nov 2022 — Jun 2023- Developed ARIMA model to predict P&L data, resulting in more accurate financial budgeting and resource allocation
- Conducted customer churn analysis of online investment platform using logistic regression, driving an increase of assets under management by 50%