Jeferson Marques is a Senior Data Scientist with eight years’ experience building production-grade fraud prevention and AI systems across fintech and public sector organizations. He has delivered high-precision identity fraud models, end-to-end ML pipelines (Kubeflow/Vertex), and automation that accelerated document processing using OpenAI and Airflow, demonstrating a pragmatic balance between immediate impact and long-term robustness. His work at Stone/Pagar.me achieved 95% precision on identity fraud while keeping false positives low, and at the Tribunal de Contas he combined BERT and tree-based models to recover millions in misclassified public spending. Fluent in Portuguese with advanced English, he codes confidently in Python, SQL and modern ML stacks (TensorFlow, PyTorch, CatBoost/XGBoost, Hugging Face) and mentors junior colleagues across cross-functional teams. Outside work he maintains sharp pattern-recognition skills through strategic online gaming, a pastime he credits for sustaining creativity and team play.
8 years of coding experience
4 years of employment as a software developer
Bacharelado, Computer Science, 8.31, Bacharelado, Computer Science, 8.31 at Instituto Federal de Goiás (IFG)
Construção de Modelos de Florestas Aleatórias para Aprendizagem de Ranqueamento através de Programação Genética
Contributions:55 commits, 30 pushes in 6 months
aleatprogramagen
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.