M9G — Minsu Chang, Ph.D. (장민수 / mincasurong): From Machine to General Intelligence by Agent-enabled System Engineering

Minsu Chang, Ph.D. (장민수 · mincasurong · M9G) is a Roboticist and Agentic AI System Engineer specializing in Vision-Language-Action (VLA) multi-robot manipulation, Human-Robot Interaction (HRI), nonlinear/adaptive control theory, assistive and rehabilitation exoskeletons (SUBAR, AI Therapist), and multi-agent coding harnesses.

Recent Highlights & Engineering Blogs

• 1st-Place Winner at Google Gemini 3 Seoul Hackathon 2026 — Created GeminiSpace, an indoor spatial AI navigation system turning room photos into interactive 2D/3D semantic maps.
• Project STAMP: Decoupling Multimodal VLA Reasoning from Robot Execution via Deterministic Pre-Dispatch Gating and Fault Write-Back (M9G-TR-2026-03)
• Building a Multi-Agent Coding Harness from Scratch and Why It Failed: Inside m9g-agent-corps vs. OmO, OpenCode, Claude Code, and Antigravity (M9G-TR-2026-02)
• Symbolic 3D Grounding and Software-Pipelined VLA Task Planning in NVIDIA Isaac Sim 6.0 & ROS 2 Jazzy (M9G-TR-2026-01)
• ROS 1/2, MoveIt 1/2 & NVIDIA Jetson Nano Camera Vision Tutorials Hub (12 Hands-On Modules)
• Full Research Archive: 11 US/KR Patents, 16 Peer-Reviewed Publications (ICRA 2025 OPPA, IEEE TNSRE AI Therapist), and KFDA Medical Exoskeletons
• LinkedIn — Minsu Chang (@mincasurong) • GitHub (@mincasurong)