Jeremy Gwinnup
Senior Research Computer Scientist at AMTA (the Association for Machine Translation in the Americas)
Dayton, Ohio, United States
Summary
🤩
Rockstar🎓
Top SchoolJeremy Gwinnup is a Senior Research Computer Scientist at AFRL’s Human Language Technologies group with over 25 years in U.S. government science and 13 years of hands-on experience in research roles, focusing on the intersection of vision and language for human-machine teaming. He adapts and evaluates generative transformer models and computer vision techniques for multimodal tasks like visual question answering, machine translation, and domain-specialized information extraction, with attention to self-hosting to protect sensitive data. Jeremy has contributed core back-end improvements to the widely used KenLM language-modeling toolkit, reflecting deep familiarity with performance-critical NLP infrastructure. He holds a Doctor of Engineering from Johns Hopkins and blends long-term DSP and ASR research heritage from AFRL with practical engineering across industry and government contractors. His work uniquely pairs foundational signal-processing expertise with modern multimodal AI to deliver applied solutions for Warfighter challenges.
12 years of coding experience
25 years of employment as a software developer
Johns Hopkins UniversityLedgemont HSM.S., Computer Science, M.S., Computer Science at Wright State University
English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (8)
c-language10
cprogramming-language10
language-modeling10
data-structure9
algorithm9
data-structures9
algorithms9
build-system8
Programming languages (8)
Github contributions (5)
kpu/kenlm
May 2015 - May 2015
Role in this project:
Back-end Developer
Back-end DeveloperContributions:34 commits, 2 PRs, 5 pushes in 3 days
Contributions summary:Jeremy primarily focused on enhancing the codebase of KenLM, a faster and smaller language model querying library. Their contributions involve incorporating command-line parsing capabilities using Boost.Program_Options, which enhances user interaction and flexibility. Furthermore, the user contributed to the offline log-linear interpolation algorithm, modifying files related to backoff reunification and adding unit tests, which indicates their involvement in improving the core functionality of the language model. These changes indicate that the user is working on core functionalities.
nlplanguage-modelsmallerkenlmfaster
jgwinnup/moses-rest
May 2016 - Aug 2019
Contributions:1 PR, 8 pushes, 1 branch in 3 years 4 months
restrest-interfacemoses
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.