🤖 Shun Ishii

Computer Vision & Physical AI Researcher / Software Developer

3D Reconstruction SLAM Physical AI Optimization

🛠 Tech Stack & Implementation Experience

Physical AI & Simulation

  • Environment configuration and physical simulation using MuJoCo (e.g., HalfCheetah, inverted pendulum tasks).
  • Validation of robot control and autonomous driving scenarios leveraging Vision-Language-Action (VLA) pipelines such as vLLM, Qwen2.5-VL, and Gemma 3.

3D Reconstruction & SLAM

  • Implementation of 3D pipelines utilizing DUSt3R / MASt3R (by NAVER LABS Europe) and VGGT-SLAM.
  • Pipeline integration converting point cloud data into STL format 3D polygon meshes via Poisson Surface Reconstruction.

Vision & Optimization

  • Object detection tasks using YOLO11 / YOLO12.
  • Implemented neutral traffic analysis logic tracking "Involved Vehicles" to avoid fault bias before signal verification.
  • Performance-focused code optimization under strict execution time (avoiding TLE) and memory limits in Kaggle.

Hardware & Infrastructure

Primary Machine: MacBook Pro (M5 Apple Silicon)

Local inference and optimization leveraging the Apple MLX framework.

Secondary Machine: Lenovo ThinkPad X1 Carbon Gen 3

Linux Mint environment / OS-level configuration and systems automation using Bash scripting.

Cloud Infrastructure: Kaggle GPU Resources

Management of on-demand GPU resources for competitions and continuous performance benchmarking.

💻 My Products & Projects

🌐 Web Applications & Tools

Vercel Deployed

Flask App

Next.js & Streamlit Apps

GitHub Pages