Daniel De Mello
Graduate Student
West Lafayette, Indiana, United States
Summary
👤
Senior🎓
Top SchoolDaniel De Mello is a CS PhD student at Purdue with nine years of hands-on experience in core machine learning research, specializing in NLP and principled analyses of transformer architectures to interpret and control how models store factual and relational knowledge. His work bridges theory and practice across deep learning topics—adversarial generative models, VAEs, Bayesian neural nets, MCMC, self-/semi-supervised learning, knowledge graphs, and time series—reflecting a breadth that enables cross-pollination of ideas. He holds an M.Sc. in Computer Science and a B.Sc. in Computer Engineering and brings a track record of exploring both model-centric and inference-driven approaches to representation and robustness. Notably, his focus on where factual memory resides in parameter space aims to make language models more interpretable and editable, an angle that informs both experimental design and potential tooling for model debugging. Located in West Lafayette, Indiana, he combines rigorous academic training with practical implementation experience cultivated over a diverse set of ML projects.
9 years of coding experience
Ph.D. Student in Computer Science, Machine Learning, Ph.D. Student in Computer Science, Machine Learning at Purdue UniversityB.Sc., Computer Engineering, B.Sc., Computer Engineering at Universidade Federal do AmazonasMaster's degree, Computer Science, Master's degree, Computer Science at Universidade Federal de Minas Gerais
Github Skills (6)
python3
machine-learning3
cudnn3
tensorflow2
neural-network2
deep-learning2
Programming languages (2)
Github contributions (5)
dmdmello/HC-MGAN
Dec 2021 - Aug 2022
Contributions:15 commits, 15 pushes, 1 branch in 7 months
dmdmello/Deep-Learning
Nov 2016 - May 2018
Deep Learning
Contributions:52 pushes, 1 branch in 1 year 5 months
pythoncaffe2deep-learningmachinelearningneural-networks
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.