top of page

WE INNOVATE DATA DRIVEN SOLUTIONS

We enable e-commerce companies to maximize customer engagement by delivering personalized and relevant content through machine learning technologies. We transform digital activities into actionable information and intelligent decisions. Our personalized recommendation systems are powered by cutting-edge machine learning technologies.

stephen-dawson-qwtCeJ5cLYs-unsplash_edited.jpg
christopher-gower-m_HRfLhgABo-unsplash.jpg

OUR MISSION

Our mission is to provide e-commerce companies, small or large, with the power of AI.  Our products empower e-commerce companies to engage and serve their customers more intelligently.

OUR TECHNICAL EXPERTISE

Our team consists of international experts in machine learning and data science.  Collectively, our accomplishments include, but are not limited to:

  • One registered US patent and 7 registered Korean patents on machine learning and deep learning

  • 10+ publications to premier AI conferences such as NeurlPS, ICML and AAAI.

  • Winner of Ischemic Stroke Lesion Segmentation Challenge sponsored by MICCAI  Society In 2016 and 2017 

1D75A9A7-CF7B-4B3C-9513-E9233A3BCB05_1_201_a.jpeg

PEOPLE

FOUNDER

MYUNGHEE CHO PAIK

CEO of Shepherd23

Professor of Statistics, Seoul National University

Formerly Professor of Biostatistics at Columbia University

PhD at University of Pittsburgh

FEATURED CO-FOUNDERS

YONGDAI KIM

Director, AI Center for Data science and Professor of Statistics 
Seoul National University
PhD in Statistics, Ohio State University


JOONG-HO (JOHANN) WON

Professor of Statistics 
Seoul National University
PhD in Electrical Engineering, Stanford University

WONCHEOL JANG

Chair and Professor, Department of Statistics
Seoul National University
Formerly, Faculty at Duke University
PhD in Statistics, Carnegie Mellon University

SUNGKYU JUNG

Associate Professor of Statistics 
Seoul National University
Formerly Faculty at U Pittsburgh
PhD in Statistics, University of North Carolina at Chapel Hill

HONGSOO KIM

Director, AI Center for Care and Health and Professor of Public Health Science
Seoul National University
Formerly Faculty at NYU
PhD in Geriatric Nursing Health System, New York University

TAESUP MOON

Associate Professor of Electrical and Computer Engineering and Graduate School of AI, AI Institute
Seoul National University
PhD in Electrical Engineering, Stanford University

JAE-KWANG KIM

LAS Dean's Professor of Statistics
Iowa State University
PhD in Statistics, Iowa State University

GI-SOO KIM

Assistant Professor of Industrial Engineering 
Ulsan National Institute of Science and Technology
PhD in Statistics, Seoul National University

YOUNG-GEUN CHOI

Assistant Professor of Statistics
Sookmyung Women's University
PhD in Statistics, Seoul National University

YONGCHAN KWON

Assistant Professor of Statistics
Columbia University
PhD in Statistics, Seoul National University

YOUNGGEUN KIM

Adjunct Associate Research Scientist of Biostatistics
Columbia University
PhD in Statistics, Seoul National University

WONYOUNG KIM

Postdoctoral Research Scientist of Industrial Engineering
Columbia University
PhD in Statistics, Seoul National University

OUR PRODUCTS

TEAM CREW

TEAM CREW is a web/app that recommends a personalized physical activity video based on a contextual bandit algorithm. This product is a prototype to collect information on user's digital behaviors and test the performance of the recommendation algorithm and identify potential problems.

PICKHOUND

PickHound is a plug-in for CAFE24, a Korean eCommerce platform for all businesses. Our plug-in provides data analytics of user interactions.  PickHound predicts individualized customer preferences in real-time and recommends items based on contextual bandit algorithms. PickHound makes high technologies accessible to e-commerce companies without internal technical teams.

PickHound

PickHound

​최첨단 AI 기술로 쇼핑몰을 성장시켜보세요!

Main(1).jpg
상품 추천이란(2).jpg
추천이 중요한 이유(3).jpg
사용해야 하는 이유(4-1).jpg
초개인화 추천(5).jpg
컨텍스추얼 벤딧(6).jpg
맞춤형 배너(7-1).jpg
무료체험(8).jpg

Thanks for submitting!

bottom of page