oryota

Hi, my name is

Ryota Okumura.

Software Engineer & Researcher

I'm a software engineer and researcher specializing in full-stack web development and deep learning. Currently pursuing my Master's degree at the University of Tokyo, focusing on computer vision and AI.

01.About Me

Hello! I'm Ryota, a graduate student at the University of Tokyo studying Information and Communication Engineering.

I'm driven by solving real-world problems through programming and turning ideas into working products quickly. Through internships and hackathons, I've gained hands-on experience across multiple engineering domains including AI/ML, data, backend, and frontend development.

I thrive in environments that demand both technical excellence and practical impact. Whether building an award-winning AI-powered LINE bot at a hackathon or developing end-to-end analytics dashboards during internships, I focus on creating robust systems that deliver real value.

Education

M.S. in Information and Communication Engineering

The University of Tokyo • 2025 - Present

Research: Semantic segmentation of artisanal small-scale gold mining using satellite imagery

B.E. in Information and Communication Engineering

The University of Tokyo • 2021 - 2025

Research: Geometric correction of AI-generated images using vanishing points

Outstanding Graduation Thesis Award (UTokyo EEIC)

Paper accepted at IEEE/CVF WACV 2026

Ryota OkumuraRyota OkumuraRyota Okumura

02.Skills

//Languages

PythonTypeScriptKotlinC/C++GoBash

//Data & Database

BigQueryDataformHasuraMySQLPostgreSQL

//Frameworks

ReactNext.jsNestJSSpring BootPyTorch

//Infrastructure

GCPDockerKubernetesTerraformLinux

03.Research

Geometric Correction of AI-Generated Images Using Vanishing Points

IEEE/CVF WACV 2026accepted
Outstanding Graduation Thesis Award (EEIC, UTokyo)

Proposed a method to detect and correct geometric inconsistencies in AI-generated images by analyzing vanishing points. The approach identifies perspective errors common in diffusion model outputs and applies corrections to improve visual realism.

PythonPyTorchComputer VisionDeep Learning

Semantic Segmentation of Artisanal Small-Scale Gold Mining from Satellite Imagery

In Progress

Developing deep learning models for detecting and segmenting artisanal small-scale gold mining (ASGM) sites from satellite imagery to support environmental monitoring and policy making.

PythonPyTorchSatellite ImagerySemantic Segmentation

04.Projects

EasyLunch

EasyLunch

AI-powered restaurant recommendation LINE bot that understands vague user preferences and suggests the most relevant restaurants based on visit history. Built at UMP-JUST Hackathon 2024.

Audience Award / Judge's Special Award

PythonLLMMySQLGCP Cloud RunLINE API
Portfolio Website

Portfolio Website

Personal portfolio website built with Next.js and Tailwind CSS, self-hosted on Kubernetes (K3s) running on Raspberry Pi 5.

Next.jsTypeScriptTailwind CSSK3sDocker

05.Intern Experience

Software Engineer Intern @ Recruit Co., Ltd.

Sep 2025 - Oct 2025

  • ▹Developed backend services in Kotlin meeting non-functional requirements
  • ▹Completed full development cycle from requirements definition to production release within 1 month
KotlinSpring Boot

06. What's Next?

Get In Touch

I'm currently looking for new opportunities. Whether you have a question, a project idea, or just want to say hi, feel free to reach out. I'll do my best to get back to you!

Say Hello