Assistant Professor · ECE · University of Arizona

SenHe

I build agentic systems for software engineering, and study how AI can test, debug, and optimize the software the world runs on.

View publications
Sen He

About

My research lies at the intersection of AI for software engineering, efficient computer vision, and human-computer interaction. I am especially interested in how large language models can be orchestrated as multi-agent systems to solve complex engineering tasks, from automated code generation to performance optimization.

Research

Four thrusts, one question: how do we let AI build better software?

01

Multi-Agent LLM Orchestration

Collaborative multi-agent architectures that leverage frontier LLMs for code generation, debugging, and refinement.

02

Software Performance Engineering

Performance-testing methodologies for cloud environments, serverless platforms, and edge/fog computing systems.

03

AI-Driven Software Testing

Harnessing LLMs to automatically test complex systems, from VR applications to cloud-native architectures.

04

Efficient Computer Vision

Applied work in medical image segmentation, brain-connectivity analysis, and 3D vision.

Selected work

Recent publications

All publications →