Gabriel Gregório

• Reduced manual analysis effort by 70% by building a production-ready NLP pipeline in Python leveraging Hugging Face BERT models and AI-driven text classification, transforming thousands of customer interactions into actionable sentiment and risk insights that improved operational monitoring and partner management.
• Partnered with Product, Operations, and Commercial teams to prioritize initi

• Increased sales by 800% by designing seller segmentation models and continuously optimizing commercial strategies through controlled experimentation.
• Improved seller retention from 30% to 90%+ by executing iterative A/B tests and refining pricing, credit, and engagement strategies.
• Generated an estimated R$15M in incremental revenue by identifying high-growth seller cohorts and designi

• Increased financing conversion rates by 15% by designing and analyzing A/B tests for pricing and lending strategies.
• Partnered with business stakeholders to translate experiment results into go/no-go strategic decisions. • Delivered data storytelling for leadership, connecting experiment outcomes to revenue impact.
• Supported decisions impacting more than R$100M in loan volume by transl

• Reduced fraud losses by an estimated 0,7% by developing anomaly detection workflows and fraud pattern identification models.
• Conducted fraud investigations using SQL and Python, performing deep exploratory analysis to validate suspicious activities and identify behavioral signatures.
• Performed large-scale case reviews and data validation processes, improving fraud detection accuracy an

• Identifying fraud patterns and designing, tracking, and monitoring fraud metrics, while implementing accurate, data-driven policies and methodologies to prevent attacks.
• Developing SQL queries to extract, transform, and aggregate data, delivering key performance insights and supporting management in the early identification of fraud trends.
• Conducting ROI analyses of fraud tools and services

I had an intense month at the Credit Recovery Department, where I was constantly challenged to bring out-of-the-box visions to the daily difficulties that the areas faced.

Altogether, I started, developed and completed three projects!

In the first, to reduce the amount of interactions with e-mail that the Back Office area needed to deal with, I partially automated Outlook for the person responsi

The course explores key Data Science methods, including Data Analysis, Information Retrieval, Data Mining, and Information Visualization. It also covers handling large-scale data (Big Data) and the latest Artificial Intelligence and Machine Learning techniques, including Deep Learning.

Curso de especialização em ciência de dados e big data