Marcel Santana
ML Research Scientist at Apple
Austin, Texas, United States
Summary
👤
Senior🎓
Top SchoolMarcel Santana is an ML research scientist based in Austin with a decade of experience applying deep learning to computer graphics, computational photography, and computer vision. He holds an M.S. in Artificial Intelligence from UFPE, where he worked on Monte Carlo denoising, and has held research roles at Apple, Intel, and Texas A&M collaborating with notable researchers. Marcel bridges research and engineering—contributing to applied projects like OpenBot by building TFRecord pipelines and augmentation-aware training workflows for robotics datasets. His background includes production-focused C++ development for visualization and biometrics, reflecting a rare blend of systems-level engineering and cutting-edge ML research. He publishes and documents work publicly (marcelsan.github.io) and brings both academic rigor and practical delivery to ML-driven imaging problems.
11 years of coding experience
9 years of employment as a software developer
Master's degree, Artificial Intelligence, Master's degree, Artificial Intelligence at Universidade Federal de Pernambuco
Portuguese, English
Github Skills (10)
data-preprocessing10
machine-learning10
tensorflow10
python10
tfrecord10
data-augmentation9
robotics8
robot8
android7
deep-learning7
Programming languages (5)
Github contributions (5)
ob-f/OpenBot
Jun 2021 - Jul 2021
OpenBot leverages smartphones as brains for low-cost robots. We have designed a small electric vehicle that costs about $50 and serves as a robot body. Our software stack for Android smartphones supports advanced robotics workloads such as person following and real-time autonomous navigation.
Role in this project:
ML Engineer
ML EngineerContributions:12 commits, 1 PR, 1 comment in 29 days
Contributions summary:Marcel focused on implementing data loading and preprocessing steps for training machine learning models within the OpenBot project. They developed a script for generating TFRecord files from the collected dataset, enabling efficient data loading. Furthermore, the user modified the training script to read data from the TFRecord format, incorporating data augmentation techniques. They also addressed issues in the training script to correctly handle the data format.
androidroboticsopenbotsmartphonerobot
marcelsan/semantic-segmentation
Jun 2018 - Nov 2018
Contributions:38 commits in 5 months
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.