Ben Peters is an NLP researcher and ML engineer with nine years of experience specializing in machine translation and multimodal speech-and-language models, currently a researcher at INESC-ID after a postdoc on the UTTER Project. He holds a PhD in Machine Learning/NLP from Instituto Superior Técnico and has published in top NLP and ML venues while teaching practical deep learning courses. Ben contributes to well-known open-source projects like OpenNMT-py, where he has fixed core NMT and PyTorch compatibility issues, reflecting strong debugging and model-engineering skills. His work spans morphologically-aware MT, noise-robust translation, and engineering for experimental psycholinguistics, combining linguistic insight with production-focused code. A Cascadian map enthusiast living in Lisbon, he brings a pragmatic, interdisciplinary approach to bridging text and speech research.
9 years of coding experience
6 years of employment as a software developer
BA, Linguistics, BA, Linguistics at University of Washington
Master of Science, Language Science and Technology, 1,3 (German system), Master of Science, Language Science and Technology, 1,3 (German system) at Saarland University
PhD, Machine Learning/Natural Language Processing, PhD, Machine Learning/Natural Language Processing at Instituto Superior Técnico
Open Source Neural Machine Translation and (Large) Language Models in PyTorch
Role in this project:
Back-end Developer & ML Engineer
Contributions:203 commits, 56 PRs, 220 comments in 1 year 7 months
Contributions summary:Ben primarily contributed to fixing bugs and improving the codebase related to neural machine translation and large language models in PyTorch. Their work included fixing Python3 compatibility issues and addressing dimension problems with MLP attention within the neural network model. They made changes to the core components of the model, as well as the translation files. These commits indicate they are proficient in debugging, model optimization, and working with neural network architectures.
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.