Rouvis Vertical AI product
AI planning workspace for new farmers in Japan, combining maps, scheduling, analytics, and conversational guidance.
Deployed prototype / ETSUZAN selected
Code / Demo / ETSUZAN feature
Bio
I am a fourth-year B.Eng. student at Nagaoka University of Technology. My work connects machine learning, industrial sensing, embedded systems, and software prototypes.
Before university, I worked at Kuroda Precision in Japanese manufacturing. I started in process development, then transferred into AIoT because of my AI and embedded systems skills, building data collection and quality-analysis workflows around real sensors and production constraints. That background shapes my current interest in AI systems that are practical, inspectable, and useful outside controlled demos.
My recent projects include vertical AI product work, manufacturing data-analysis tools, automotive hardware architecture exposure, and B.Eng. research on reliability-aware KV-cache compression for long-context LLM inference under Prof. Akiko MANADA.
Research Interests
- KV-cache compression for LLMs. My B.Eng. research under Prof. Akiko MANADA studies reliability-aware KV-cache compression for long-context LLM inference, focusing on how to preserve structurally important tokens under memory constraints.
- Applied AI systems. I am interested in AI systems that move beyond demos and account for data quality, deployment constraints, and user workflows.
- Sensing and AIoT. I have worked with sensor-integrated tooling, production data, and analysis workflows in manufacturing environments.
- Simulation and visualization. I use interactive tools and visualization to make technical behavior easier to inspect, explain, and improve.
Selected Work
Interactive Torque Project Industrial tooling
Torque measurement visualization and analysis tool for industrial inspection data and pass/fail evaluation.
Analysis application
Ceramic Fracture Lab Simulation / research
Interactive ceramic 3-point bending simulations for teaching, rapid prototyping, and early-stage fracture studies.
Research prototype
Firebase ESP32 CT Monitor Embedded / IoT
ESP32-based current transformer monitoring pipeline with Firebase-backed remote data access.
Embedded prototype
Experience
- Aug 2025 to Sep 2025 HW Architect Intern at Bosch Japan
Worked on automotive E/E architecture and ADAS hardware platform customer-interface work in a hybrid internship setting.
- Jul 2025 to Aug 2025 AI Engineer Intern at OPTiM
Implemented Fusion RAG system work for the agentic AI chat product OPTiM AIRES.
- 2025 to 2027 B.Eng. candidate at Nagaoka University of Technology
Studying electrical, electronics, and information engineering under Prof. Akiko MANADA, with B.Eng. research on reliability-aware KV-cache compression for long-context LLM inference.
- 2025 to present Freelance manufacturing AI developer
Delivered image-analysis and vibration-classification tools for manufacturing clients after leaving full-time industry work.
- May 2023 to Apr 2025 AIoT Engineer at Kuroda Precision
Transferred into AIoT work because of AI and embedded systems skills, then built sensor data pipelines, quality analytics, and machine-learning workflows for production environments.
- Apr 2022 to Apr 2023 Process Development Engineer at Kuroda Precision
Worked on electric-vehicle motor-core manufacturing process development before an internal transfer into industrial IoT engineering.
Personal
Outside engineering, I enjoy games like Dota 2, espresso, and long-distance running. I have completed several full marathons in Japan.
- Languages. Thai (native), English (C2), Japanese (JLPT N1).
Contact
For research, internship, engineering, or collaboration inquiries, email is the best contact path.