About

I am a software engineer and AI systems researcher currently working as a Research Assistant at the University of Virginia. My work focuses on large language models, adaptive generation, representation learning, and AI systems. I am currently profiling vLLM-based agentic workloads to study concurrency, KV-cache management, and serving efficiency. More broadly, I am interested in how language models can adapt their behavior at inference time and how internal representations can support that adaptation.

Before joining UVA, I conducted independent LLM research. Together with Muyao Kong, I co-developed Scenario-Actor, a scenario-adaptive decoding framework that learns representations from Llama-3.1-8B-Instruct hidden states and uses them to select prompt-level sampling temperatures. Our paper, Read the Room Before Your Model Generates, was accepted to the AACL-IJCNLP 2026 Main Conference.

Previously, I spent over three years as a Software Development Engineer at Amazon Web Services, working on Amazon Connect. I built and operated large-scale distributed systems supporting more than 20 million daily customer interactions, and worked on retrieval-grounded, tool-integrated LLM workflows for production backend systems. That experience strongly influences how I approach research: I care about methods that remain practical at scale and evaluations that clearly establish what a system can and cannot do.

I received an M.S. in Computer Science from Santa Clara University and a B.S. in Computer Science, cum laude, from the University at Buffalo.

News

• Oct 2026 — Our AACL-IJCNLP 2026 paper is now on arXiv: Read the Room Before Your Model Generates.
• Sep 2026 — Started as an AI Systems Research Assistant at the University of Virginia, working on efficient inference for agentic LLM workloads.
• Sep 2026 — Our paper “Read the Room Before Your Model Generates” was accepted to the AACL-IJCNLP 2026 Main Conference.
• May 2026 — Joined Binhao Technology as an Applied AI Engineer, building AI-assisted automation for manufacturing workflows.
• Feb 2023 — Returned to Amazon Web Services as a full-time Software Development Engineer on the Amazon Connect Analytics team.
• Jun 2022 — Joined Amazon Web Services as a Software Development Engineer Intern on the Amazon Connect Analytics team.
• Jan 2022 — Started as a Research Assistant at Santa Clara University, working on NLP research with Prof. Yi Fang.