Curriculum Vitae

Qing Zhong

Boston, MA | +1 857‑361‑8333 | zhong.qing@northeastern.edu | GitHub

Neuroergonomics and Physiological Sensing; Human–AI Interaction;Human-Centered Design & Evaluation.

Tufts University 2023/01 – 2025/05

M.S. in Computer Science Boston, MA

Northeastern University 2020/09 – 2022/08

B.S. in Information Technology Boston, MA

LLM‑Based Tools and Their Effects on Cognitive Workload in Decision Making 2024/05 – Present

Graduate Research Assistant, Tufts University HCI Lab

Advisor: Prof. Robert Jacob

Investigated how LLM-based tools (Microsoft Copilot) affect human cognitive workload in decision-making, measured via fNIRS.
Conducted experiment design and ran participant sessions (n=32) to collect fNIRS data.
Developed Python-based pipelines for processing fNIRS data and feature extraction and applied linear mixed‑effects models to compare cortical activation across AI‑assisted vs. control conditions.
Found reduced workload for novice users under AI assistance; increased workload for novice users who rely on AI early, while experienced users showed stable workload.
Contributed results to a peer‑reviewed extended abstract at the Neuroadaptive Technology Conference (NAT’25).

LLM Browser Assistant for Accessible E‑Commerce for the Elderly 2025/01 – 2025/05

Team Lead, Course‑Based Research Project (HCI for Disability)

Advisor: Prof. Elaine Short

Designed and implemented a browser extension using LLMs to support older adults in navigating complex e-commerce websites.
Conducted a within-subjects user study (n=9), measuring task performance and mental workload (NASA-TLX).
Found higher task success rates, reduced frustration, and improved self-rated performance under the LLM assistant condition.
Derived design guidelines for AI-driven accessibility tools and prepared an extended manuscript for submission.

LLM‑Tools' Effects on Users During Complex Decision‑Making with fNIRS (Extended Abstract). [PDF]
M. Russell, Q. Zhong, K. Zheng, K. Hu, J. Santaniello, R. J. K. Jacob. Neuroadaptive Technology Conference (NAT'2025)

Medical LLM Reasoning 2025/05 – 2025/06

Fine-tuned Meta‑Llama‑3.1‑8B‑Instruct on a medical QA dataset using LoRA and 8‑bit quantization.
Improved accuracy on a clinical QA dataset, demonstrating the potential of domain-adapted LLMs for reliable healthcare applications.

Medsafe AI Agent 2025/01 – 2025/03

Designed a multi-agent clinical decision-support system with dynamic tool selection and 4-stage autonomous workflow for drug interaction analysis.
Achieved 35% improvement in answer completeness through hybrid retrieval and self-correction, demonstrating effective agentic reasoning for trustworthy healthcare AI.

Scholar AI RAG System 2024/10 – 2025/01

Built a multimodal RAG system with OCR-based layout parsing and hybrid retrieval for academic literature discovery with automatic citation grounding.
Achieved Top-5 recall 65% to 88% and MRR 0.50 to 0.72, with high answer traceability; highlighting explainable retrieval as key for reproducible research.

Software Engineer Intern, Chengdu Knowing Link Tech 2023/05 – 2023/09

Developed scalable backend services for a gaming platform (Flask/MySQL/Redis).
Applied caching & query optimization to reduce latency and enhance system reliability.

Software Engineer Intern, Welly's Wellness 2022/03 – 2022/06

AI & ML: Machine Learning, Large Language Models, AI Agents, PyTorch, Fine-tuning.
Research: Experimental design, User studies, Usability testing, Survey design, Statistical analysis, Signal processing.
Programming & Tools: Python, Java, FastAPI, Flask, Spring Boot, SQL, Docker, Huggingface, LangChain, LlamaIndex, Git, Latex.