Takuma Yamaguchi
VP Of Engineering, Mercari US at Mercari, Inc.
Chiyoda, Japan
Summary
👤
Senior🎓
Top SchoolTakuma Yamaguchi is a VP of Engineering leading Mercari US with over a decade of hands-on experience in machine learning, computer vision, big data and backend systems. He progressed from academic researcher (Ph.D. in Information Engineering) to building production ML/search/data platforms and now drives engineering strategy and operations across Mercari’s U.S. organization. His technical breadth spans TensorFlow, Hadoop/Spark/Hive/Presto, vector and semantic search, microservices on Kubernetes, and production languages from Python to C++, enabling end-to-end delivery from research prototypes to scalable services. He has repeatedly combined CV research—publishing and shipping image-recognition features and visual search—with product-focused engineering such as automated moderation, listing UX automation, and KPI-driven growth experimentation. An early contributor to an image search prototype integrating VGG16 and Flask for fast similarity retrieval, he uniquely blends deep academic roots with pragmatic system engineering at marketplace scale. Based in Chiyoda, Japan, he is known for turning complex ML research into measurable product outcomes.
9 years of coding experience
15 years of employment as a software developer
Ph.D. in Engineering, Information Engineering, Ph.D. in Engineering, Information Engineering at Shinshu University
Japanese, English
Github Skills (9)
keras10
python10
vggnet10
flask10
image-retrieval10
flask-ask10
numpy9
tensorflow9
computer-vision8
Programming languages (10)
Github contributions (5)
matsui528/sis
Mar 2017 - Jul 2021
Simple image search engine
Role in this project:
ML Engineer
ML EngineerContributions:28 commits, 6 PRs, 40 pushes in 4 years 4 months
Contributions summary:Takuma primarily contributed to the development of an image search engine. Their work involved integrating the VGG16 model for feature extraction, creating a Flask-based web application to handle image uploads, and performing similarity searches. They also implemented an offline script to precompute image features and store them for faster retrieval, along with updating the frontend with a basic HTML template.
search-enginespythonflasksearch-engineimage-retrieval
mti-lab/math_writing
Jan 2021 - Jan 2023
Introduction to mathematical writing for undergraduate students majoring in engineering
Contributions:1 release, 12 commits, 10 PRs in 2 years
undergraduatemathmathematicsmathematicalengineering
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.