UC Berkeley · MEng 2027

Guanchen Liu.

Machine Learning & Computer Vision Engineer

I build machine learning systems for real-world engineering problems — from visual inspection robots to graph-based transportation models and optimization tools.

Portrait of Guanchen Liu

About Me

I'm a Master of Engineering student in Systems Engineering at the University of California, Berkeley. My work combines deep learning, computer vision, optimization, and engineering applications. I enjoy turning research ideas into working prototypes and measurable results.

Selected Projects

CloudJet inspection robot prototype

CloudJet Robotics — Climbing Inspection Robot

Led a nine-person team developing a steel-structure climbing inspection prototype, integrating sensing, corrosion segmentation, and remote control. Prototype climbing performance was tested.

PythonPyTorchOpenCVU-NetRoboticsEmbedded Systems
View project & demo ↗

Traffic Assignment GNN Surrogate

Built a graph neural network surrogate for traffic flow prediction on the public Sioux Falls network, exploring faster alternatives to iterative assignment.

GNNPyTorchTransportation
Ask about this project ↗

Maritime Resource Optimization

Built a simulation-driven resource optimization tool using OR-Tools CP-SAT and a Flask dashboard. Team awarded third place at a maritime hackathon.

OptimizationCP-SATFlask
Explore GitHub ↗

Technical Skills

Programming

PythonMATLABC++SQLJavaScript

Machine Learning & Computer Vision

PyTorchOpenCVCNNU-NetResNetCOLMAPTensorFlowscikit-learnTransformerLLMGNN

Engineering & Tools

FastAPIFlaskpytestGitGitHub ActionsCUDACloud GPU ComputingOR-ToolsSignal Processing

Publications

Patents

Let's Connect

Interested in machine learning, computer vision, robotics, or applied AI? Feel free to reach out.