Felipe Marcelino
Data Scientist
Greater Belo Horizonte Brazil
Summary
👤
Senior🎓
Top SchoolFelipe Marcelino is a data scientist with a decade of professional experience and over six years focused on machine learning, statistical modeling, and production-grade data systems. He has delivered measurable impact across finance (fraud detection), retail (demand and pricing), industry 4.0 (steel quality), healthcare (computer vision for glaucoma), and energy/resource allocation using reinforcement learning. Comfortable in both Python and Scala ecosystems, Felipe builds end-to-end solutions—APIs, Kubeflow pipelines, and scalable Spark processing—that bridge research and production. At Nubank he improved fraud detection by 33% using graph features, and his consulting work applied RL and boosting models to complex operational problems like ship allocation and nationwide SKU forecasting. He holds a Master’s in Computer Science and pairs strong stakeholder communication with practical XAI to make models actionable for domain experts. Based in Greater Belo Horizonte, he combines academic rigor with hands-on software engineering to turn ambiguous business questions into deployed AI products.
10 years of coding experience
9 years of employment as a software developer
Master's degree Computer Science, Master's degree Computer Science at Universidade Federal de Minas Gerais
Portuguese, English
Github Skills (141)
toolbox10
powerline10
windows10
statusline10
material-design-icons10
font-size10
stable-baselines10
deep-learning10
iconic10
todoist10
robotics10
heroicons10
aggregator10
linux10
gradient-boosting10
Programming languages (13)
Github contributions (5)
FelipeMarcelino/nix-dotfiles
Oct 2023 - Mar 2024
My NixOS/Darwin/Flakes/Home Manager configuration for everything!
Contributions:77 PRs, 137 pushes, 51 branches in 4 months
configurationdotfileshome-manageri3wmmacos
FelipeMarcelino/2048-Gym
Mar 2020 - Aug 2020
This projects aims to use reinforcement learning algorithms to play the game 2048.
Contributions:12 commits, 25 pushes, 1 branch in 4 months
ddqnreinforcement-learning-algorithmsmultiagent-reinforcement-learningreinforcement-learningmahjong
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.