Gish Shao
Computer Vision Engineer at TREVI 特雷維科技
New Taipei, Taiwan
Summary
👤
Senior🎓
Top SchoolGish Shao is a Computer Vision Engineer with 10 years of experience building low-latency, real-time inference systems and end-to-end ML pipelines, currently focusing on edge deployment and streaming applications at TREVI. He specializes in card-game recognition and robust E2E integration—self-collecting training data, authoring annotation tools, and achieving >99% online accuracy in production pipelines. Gish has a strong DevOps and inference optimization background, migrating legacy TensorFlow 1 models to TensorFlow 2, designing FastAPI socket gateways, and creating OBS-compatible mock systems to stress-test streaming scenarios. His work spans OCR and NLP in production claim-processing systems, where he boosted receipt OCR accuracy from 80% to 96% by integrating Azure Form Recognizer and tailored postprocessing. An experimental researcher too, he benchmarks 7B-scale LLMs for real-time decision advisors and publishes related RAG experiments and demo code on GitHub. Based in New Taipei, he combines business-school rigor with hands-on ML engineering to turn messy real-world data into reliable, maintainable inference services.
10 years of coding experience
4 years of employment as a software developer
Bachelor's degree, Major : Management Information System and Minor : Accounting, 3.75 GPA, Bachelor's degree, Major : Management Information System and Minor : Accounting, 3.75 GPA at 國立政治大學Master's degree, Business Administration, Master's degree, Business Administration at 國立臺灣大學
English, Chinese
Github Skills (40)
captcha9
interpretable-machine-learning9
blackbox9
iml8
transparency8
navigator8
xai8
shap7
binning7
anti-bot7
selenium7
testing7
imperva7
selenium-webdriver7
shapley7
Programming languages (3)
Github contributions (5)
GISH123/Cathay-Holdings-CIP-Projects-for-Interpretable-Machine-Learning
Aug 2019 - Sep 2019
Contributions:60 commits, 59 pushes, 1 branch in 27 days
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.