Jasper Xian
Research Engineer at Google DeepMind
New York City Metropolitan Area United States
Summary
🤩
Rockstar🎓
Top SchoolJasper Xian is a Research Engineer at Google DeepMind with five years of hands-on experience building and researching large language models and distributed training systems. He has interned across top AI labs—including Cohere and DeepMind—working on distributed LLM training, evolutionary algorithms for LLMs, and product-focused ML systems at Wish. Jasper contributes to open-source ML tooling (notably integrating Pyserini retrieval into Stanford NLP’s DSPy framework), reflecting a practical focus on retrieval-augmented model workflows and reproducible research. Based in the NYC metro area and trained at the University of Waterloo, he blends foundational research interests with production-minded engineering, and describes himself on GitHub with a playful curiosity as an "aspiring ball knower."
5 years of coding experience
High School Diploma, High School Diploma at West Windsor-Plainsboro High School NorthBachelor of Computer Science - BCS, Computer Science, Bachelor of Computer Science - BCS, Computer Science at University of Waterloo
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (7)
machine-learning10
dnspy10
faiss10
python10
google-colaboratory9
nlp9
google-colab9
Programming languages (3)
Github contributions (5)
stanfordnlp/dspy
Sep 2023 - Dec 2024
DSPy: The framework for programming—not prompting—language models
Role in this project:
ML Engineer
ML EngineerContributions:4 reviews, 4 PRs, 7 pushes in 1 year 3 months
Contributions summary:Jasper primarily contributed to integrating the Pyserini retrieval system into the DSPy framework. Their commits involved modifying existing code to incorporate Pyserini for document retrieval using prebuilt and local indexes. They also made adjustments to the codebase for Colab compatibility and improved documentation. The changes demonstrate a focus on enhancing the retrieval capabilities of DSPy.
nlpbertknowledgepredictlanguage-models
jasper-xian/pyserini
Mar 2022 - Jan 2025
Pyserini is a Python toolkit for reproducible information retrieval research with sparse and dense representations.
Contributions:2 PRs, 127 pushes, 42 branches in 2 years 10 months
semantic-similaritypythonrepresentationsbrain-computer-interfacesparse
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.