Yen-chun Chen
Member Of Engineering at poolside
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolYen-chun Chen is an ML researcher-engineer with a decade of experience building and scaling language and multimodal models, currently working on pretraining-scale token synthesis at Poolside to advance AGI. Previously a researcher at Microsoft, he contributed to cutting-edge projects like Phi3-Vision and Phi-4 multimodal and reasoning models and has a strong grounding in coding LLMs. His open-source work includes implementing and optimizing abstractive summarization models (ACL 2018 codebase) with sequence-to-sequence, copy mechanisms, and efficient data pipelines. Trained with an MS in Computer Science from UNC and a BS from National Taiwan University, he blends academic rigor with production-focused engineering. An understated strength is his ability to move research prototypes into performant systems, evidenced by both corporate model releases and hands-on repo contributions.
10 years of coding experience
8 years of employment as a software developer
Bachelor of Science (BS) Electrical Electronics and Communications Engineering, Bachelor of Science (BS) Electrical Electronics and Communications Engineering at National Taiwan UniversityMaster of Science (MS) Computer Science, Master of Science (MS) Computer Science at University of North Carolina at Chapel Hill
English, Chinese
Github Skills (11)
attention-mechanism10
sequence-to-sequence10
pytorch10
machine-learning10
nlp10
word2vec10
python10
data-preprocessing9
python-multiprocessing8
multiprocessing8
multi-process8
Programming languages (2)
Github contributions (5)
ChenRocks/fast_abs_rl
May 2018 - Apr 2019
Code for ACL 2018 paper: "Fast Abstractive Summarization with Reinforce-Selected Sentence Rewriting. Chen and Bansal"
Role in this project:
ML Engineer
ML EngineerContributions:76 commits, 2 PRs, 9 pushes in 11 months
Contributions summary:Yen-chun implemented and refined machine learning models for abstractive summarization. Their commits focus on pretraining a word2vec model, creating datasets for extraction labels, and developing a sequence-to-sequence model with attention. They also optimized data processing with multi-processing techniques and incorporated a copy mechanism into the summarization model.
nlpsentenceabstractive-summarizationchenrewriting
ChenRocks/UNITER
Jan 2020 - Jun 2021
Research code for ECCV 2020 paper "UNITER: UNiversal Image-TExt Representation Learning"
Contributions:49 commits, 14 pushes, 16 comments in 1 year 5 months
pytorchimage-textrepresentationdeep-learningeccv
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.