Ching-Hsiang (Tom) Hsu

I am a senior software engineer working at the intersection of AI infrastructure, robotics, and real-time systems. My work focuses on the software layer behind modern intelligent products: low-latency C++ systems, deterministic runtime, validation and replay infrastructure, performance profiling, and system integration for complex autonomous and embedded platforms.

I am especially interested in the engineering foundation that makes AI usable in production, including runtime reliability, observability, multi-threaded performance, and the tooling needed to connect models and algorithms with real-world products. Across autonomous driving and embedded software, I have worked on the layers that sit between research and deployment, where system behavior, performance constraints, and product requirements all have to come together in a practical way.

Overview

Core Stack

C/C++, Python, Java, HTML, Matlab
Docker, Git, Protobuf, OpenCV, PCL, Qt, OpenGL

Focus Areas

AI infrastructure, real-time systems, robotics software, motion planning, routing and navigation, localization and mapping

Engineering Style

System thinking, cross-functional execution, operational rigor, code review discipline, and translating research ideas into production-ready architecture

Languages

English, Chinese, Taiwanese

Selected Highlights

Autonomous Driving Leadership

Led multidisciplinary development spanning planning, control, integration, and infrastructure for autonomous driving systems operating in complex urban environments.

Production-Grade Systems

Built and improved real-time software with strong emphasis on safety handling, failure recovery, simulation, deployment readiness, and system observability.

Research to Product

My background combines robotics research, geometric algorithms, perception, and software engineering, which helps me bridge theory, implementation, and product impact.

Continuous Learning

I actively document hands-on explorations in modern C++ and open-source AI systems, with a particular interest in embodied intelligence and infrastructure for intelligent agents.