Summary
Yasutaka Shindoh is a data scientist based in Fukuoka with 11 years of experience applying information retrieval, NLP, and language-model techniques to product search, advertising, and speech systems. He has led end-to-end projects that moved ranking, query understanding, autocomplete, faceted search personalization, and attribute extraction from research into production. His background spans industrial R&D and engineering roles—from speech recognition and voice interfaces to corpus creation and offensive-content detection—anchored by a Ph.D. in Mathematics. He combines rigorous academic training with practical system-building, including tokenizer and dictionary enhancement for product search. Colleagues benefit from his ability to translate language-model research into deployable features that improve user search and recommendation quality. He often works at the intersection of query intent modeling and applied linguistics, a less visible strength that drives measurable search relevance gains.
11 years of coding experience
21 years of employment as a software developer
Kyoto University
Doctor of Philosophy (Ph.D.), Mathematics, Doctor of Philosophy (Ph.D.), Mathematics at Osaka City University