Summary
Hiroki Tanabe is a data scientist with seven years of experience who combines rigorous academic training—a Ph.D. in computer science from Kyoto University—and hands-on ML engineering to build CVR prediction models for high-throughput advertising platforms. At LINEヤフー he owns the full model lifecycle, from PySpark data aggregation and TensorFlow/LightGBM training to deployment with TensorFlow Serving/Triton and MLOps automation via Airflow and Argo. His research background in multiobjective optimization produced published algorithms and an open-source Python package, reflecting his ability to translate theory into production-ready code. He has a track record across startups and industry (HACARUS, Cookpad, Yahoo) solving image, recommendation, and causal inference problems, and he complements his technical work by shipping monitoring tools like a Streamlit dashboard. Outside of work he runs two YouTube channels and builds a Python-driven AI VTuber pipeline that autonomously generates and uploads news commentary, demonstrating a rare blend of creative product building and systems automation.
7 years of coding experience
1 year of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Kyoto University
English, Japanese