Kohei Yoshikawa has been a Research Engineer at NTT DATA Mathematical Systems Inc., Japan, since April 2021, and a PhD student at the Graduate School of Mathematics, Kyushu University, Japan, since April 2024.
He received B. Eng. and M. Eng. degrees from The University of Electro-Communications, Japan, in 2019 and 2021, respectively. Since April 2021, he has been in his current position.
Last update: 2025.4.23
Research Engineer, Data Mining Division
Apr. 2021 – Apr. 2024
Chief Research Engineer, Data Mining Division
Apr. 2024 – Present.
Designed, implemented, and maintained core statistical and machine‑learning libraries in Data Analytics software.
Led front‑end architecture and UX design for the BayoLinkS v9.2.0 release (March 28, 2024), including the “Causal Mapping” icon that enables intuitive diagram‑based constraint setting for Bayesian network structure learning.
Enhanced the Causal Mapping feature with capabilities such as free‑form diagram drawing, data‑driven variable placement, style customization, group creation, and direct export of causality constraints for downstream analysis.
Adopted advanced UX design methodologies—including mental model diagrams, user scenarios, and object‑oriented UI (OOUI)—to align stakeholder vision, streamline feature ideation, and accelerate delivery cycles.
I have consistently managed the entire data analysis process—organizing client requirements, planning the analysis, executing it, summarizing results, and delivering comprehensive reports. This end-to-end involvement has enabled clients to make informed decisions and contributed to their business success.
I also provide consulting services to clients facing challenges in data analysis, offering tailored analytical approaches and methodologies that lead to more effective and insightful outcomes.
Developed and delivered comprehensive statistics and machine learning training programs, equipping students and professionals with practical data‑science skills.
Proof of Concept (PoC) for Generative AI
Project Leadership & Technical Lead: Managed end-to-end PoC engagements—scoping with clients, defining objectives, and delivering demonstrators that showcase business value.
Azure-based AI Validation: Deployed and configured Azure environments to conduct prompt engineering and Retrieval-Augmented Generation (RAG) experiments.
Prototype Design & Delivery: Architected, developed, and handed over prototype applications integrating generative AI models.
Value Creation Commitment: Collaborated directly with client stakeholders to translate business challenges into technical requirements and ensured PoC outcomes aligned with their strategic goals.
Apr., 2024 – (Present)
Graduate School of Mathematics,
Kyushu University, Japan.
Advisor: Prof. Shuichi Kawano
Apr., 2019 – Mar., 2021
Department of Computer and Network Engineering, Graduate School of Informatics and Engineering,
The University of Electro-Communications, Japan.
Degree: Master of Engineering (as valedictorian in the department)
Advisor: Assoc. Prof. Shuichi Kawano
Dissertation: Tensor Common Component Analysis based on Kronecker Product Representation
Apr., 2015 – Mar., 2019
Department of Communication Engineering and Informatics, Faculty of Informatics and Engineering,
The University of Electro-Communications, Japan.
Degree: Bachelor of Engineering
Advisor: Assoc. Prof. Shuichi Kawano
Dissertation: リーマン多様体上における多変量低ランク回帰モデルの推定
Apr., 2012 – Mar., 2015
Tokyo Metropolitan Kunitachi Senior High School (東京都立国立高等学校)
Statistical machine learning, Sparse modeling, Dimension Reduction, Causal Inference.