Xin Xu
PhD Candidate
Summary
🤩
RockstarXin Xu is a research scientist and machine learning engineer with seven years of experience bridging academic research and production-grade systems, currently building low-latency, high-throughput LLM inference infrastructure at ByteDance in California. Trained at Peking University and UC San Diego/Illinois (MS in Computer Science), Xin has a strong research footprint in video understanding, diffusion models for perception, open-vocabulary segmentation, and efficient high-resolution image understanding from internships at Microsoft Research and Adobe. He combines deep learning research with systems-level thinking—optimizing model inference performance as well as model accuracy—making him effective at moving ideas from paper to deployable platforms. Xin’s background in teaching discrete structures and computational photography reflects a penchant for clear technical communication and foundational rigor. Notably, he has worked on disentanglement for video compression and temporal dynamics pooling for action recognition, indicating a consistent focus on spatiotemporal representation learning across modalities.
7 years of coding experience
Github Skills (8)
transformers10
pytorch10
nlp10
deep-learning10
python10
bert10
named-entity-recognition9
ch8
Programming languages (3)
Github contributions (5)
zjunlp/LREBench
Jun 2022 - Jan 2023
Contributions:54 commits, 50 pushes in 7 months
xxupiano/ChineseMusicTransformer
May 2021 - Dec 2021
Contributions:22 commits, 21 pushes, 1 branch in 7 months
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.