• Built an automated evaluation framework for multi-turn LLM conversations, applying automatic prompt optimization for foundation models to develop LLM-as-a-Judge evaluations of question quality, logical flow, hallucination, and policy compliance, standardizing metrics across 5M+ customer–LLM dialogues per day.

• Designed and deployed Responsible AI guardrails for return-related conversations, implementing safeguards against misuse, prompt injection, and content violations; optimized for real-time inference with 50+ transactions per second (TPS) to ensure fast and cost-efficient performance at production scale.

• Developed an evaluation pipeline for customer comments and LLM-generated follow-up questions, using RAG-based retrieval and prompt engineering to measure information gain in responses and assess question relevance, clarity, and safety; eliminated manual review, saving over $500K and 25,000+ hours of labeling annually.

• Extended diffusion-model research into a framework for safeguarding customer images against generative misuse, enabling practical image protection in Amazon’s Responsible AI ecosystem; recognized through a WACV 2025 publication, a pending patent, and the Amazon Inventor Award for innovation and impact.

• Developed a novel and practical defense method to protect user images in real-time from identity theft and copyright misuse in generative diffusion models by introducing latent-space adversarial perturbations.

• Engineered an optimized perturbation generation and inference framework that accelerated processing speed by 150x (from ~30s to 0.2s per image) and reduced memory usage by 2.6x while maintaining protection performance.

• Evaluated the system on three public diffusion benchmarks and 10 000+ real-world samples using over 10 evaluation metrics, demonstrating superior protection strength, visual fidelity, and generalization with up to 20% improvement over existing diffusion-model protection methods.

• Conducted leading-edge research in Deep Learning, identifying emerging research challenges and opportunities in the field. Facilitated research direction and idea exchange by hosting a weekly paper reading group.

• Designed and executed extensive experiments to validate hypotheses, resulting in 10 published papers at top AI conferences (NeurIPS, ECCV, AAAI, ICASSP, MobiCom). Effectively communicated research findings through concise, well-written academic papers and conference presentations.

• Supported course instructors by organizing weekly class recitations, grading assignments, and holding office hours. Mentored and assisted over 200 students in foundational and advanced Computer Engineering courses.

• Developed an end-to-end machine learning pipeline for data cleaning, labeling, and DNN training, enabling accurate ID card information extraction and powering an in-house ID verification system used by 10M+ users.

• Designed and deployed a secure internal data-labeling platform integrated into the pipeline, processing 20 000+ high-quality annotations and reducing external labeling costs by more than $100 K.

• Enhanced OCR model performance by 10% through targeted data curation and model fine-tuning, delivering production-level accuracy and reliability in large-scale deployment.

• Installed, upgraded, maintained, and ensured smooth operation of over 100 PCs, Macs, printers, projectors, and mobile devices.

• Provided top-notch customer support and resolved client issues in person, via phone, and remote desktop connections for more than 100 staff and faculties with a wide range of technical backgrounds.

• Performed diagnostics and troubleshooting system issues, documented help desk tickets/resolutions, and maintained equipment inventory lists.

• Planned a 7-day-long orientation program for more than 2000 international students.

• Helped more than 1000 new students integrate into American academic culture via social events.

• [CVPR-23, 24] IEEE / CVF Computer Vision and Pattern Recognition Conference

• [ECCV-22, 24] European Conference on Computer Vision

• [NeurIPS 23, 24] Conference on Neural Information Processing Systems.

• [ISCAS-23] IEEE International Symposium on Circuits and Systems

• [AICAS-22] IEEE International Conference on Artificial Intelligence Circuits and Systems

Rutgers University, New Brunswick, NJ, USA

• [Fall 2022] Digital Logic Design

• [Spring 2022] Probability and Random Processes

• [Spring 2024] Introduction to Deep Learning