Virginia Tech

College of Engineering
PhD candidate in Computer Science

University of North Carolina at Chapel Hill

Gillings School of Public Health
Master of Science in Biostatistics

Zhejiang University

School of Mathematical Sciences
Bachelor of Science in Statistics

Research Intern

Microsoft, Redmond, USA

• • Lead research on RL-based post-training and test-time search to improve the reliability and efficiency of LLMs in agentic Copilot systems, while designing and developing tool-augmented LLM agent pipelines in live environments to enable multi-LLM collaboration and long-horizon tool-use workflows.

Graduate Research Assistant

Computer Science, Virginia Tech, Blacksburg, USA

• • Developed scalable methods to improve the reliability and efficiency of frontier AI systems, with a focus on hallucination mitigation, uncertainty quantification, and adaptive test-time efficiency—advancing the theoretical and empirical foundations of assured, efficient, and trustworthy large language models in real-world deployments.

Visiting Scholar

Computer Science, Virginia Tech, Blacksburg, USA

• • Conducted research at the intersection of statistical learning theory and foundation models, developing principled probabilistic frameworks to characterize LLM generalization and reliability—laying the theoretical groundwork for subsequent work on model selection and uncertainty quantification.

Graduate Research Assistant

Biostatistics, UNC-Chapel Hill, Chapel Hill, USA

• • Developed scalable Bayesian probabilistic frameworks and uncertainty-aware ML models for large-scale data inference, focusing on principled statistical learning methods with applications to heterogeneous, high-dimensional datasets.

Student Researcher

Statistics, Zhejiang University, Hangzhou, China

• • Applied statistical optimization and deep learning to build state-of-the-art predictive systems in quantitative finance, developing end-to-end ML pipelines with measurable improvements in forecasting accuracy and computational efficiency.

Software Design Engineer Intern

Alibaba, Hangzhou, China

• • Redesigned a legacy static-path system with a Java-based Dynamic Source Routing (DSR) algorithm incorporating a node-energy score, boosting end-to-end packet delivery from 88% to 97% and reducing average latency by 30% in 500-node stress tests under highly dynamic traffic.