• University: San Diego State University
• College: NA
• Degree: Master of Science (MS)
• Major: NA
• Minors: NA
• Start Date: NA

Ai Research Associate

• Developed a multimodal preeclampsia risk pipeline fusing retinal vascular biomarkers with clinical vitals across 393 patient records and 57 engineered features.
• Benchmarked Logistic Regression, Random Forest, XGBoost, and CatBoost over 162 hyperparameter configurations; CatBoost reached 92.4% accuracy and 0.957 ROC AUC, a 31% AUC improvement over baseline.
• Engineered visit-level feature fusion from 4 retinal quadrant images per patient into structured, model-ready features; delivered a CatBoost screening model with 92.9% precision and 0.81 F1.

Research Intern

• Reduced large-language-model inference power consumption by 4.5% using EasyQuant post-training quantization, eliminating an estimated $1,200/year in cloud compute costs per deployed instance.
• Instrumented GPU energy telemetry across 6 hardware configurations; cut average inference latency by 18% and saved 40+ GPU-hours/week in multimodal workload processing.
• Ranked the top quantization strategy among 4 competing approaches through systematic benchmarking; the selected approach became the team standard and supported a peer-reviewed paper.

• LGBTQ: NA
• Ethnicity: NA
• Disability: NA
• Veteran: NA
• First-generation College Student: NA
• Pell Grant Eligibility: NA
• Work Authorization: NA
• Visa Sponsorship: NA