Predicting Customer Satisfaction

Understanding customer satisfaction is difficult because feedback is scarce. I developed an ensemble of machine learning models to predict customer satisfaction based on behavioral signals, which can be used to enhance downstream systems including product ranking and behavior modeling.

Ranking Products with LLMs

Learning to rank products is a complex task that requires understanding user intent and product relevance. I developed a listwise-ranking LLM that demonstrated superior performance over a production baseline in key ranking metrics, significantly improving search relevance.

Deep Learning for Radiology

Medical imaging presents unique challenges such as limited labeled data, class imbalance, label noise, and device-specific variability. I contributed to the development and analysis of deep learning models across diverse modalities including chest X-ray, CT, MRI, and histopathology.

Promoted to Research Lead

Initiated long-term research directions across teams, collaborated closely to drive publications, and mentored junior researchers.

Published Research in AI & Medical Venues

Authored first-author papers at machine learning conferences (NeurIPS, AAAI) and contributed to multiple publications in leading medical imaging venues like MICCAI.

Cryptography in Embedded Systems

Developed a low-latency, minimal-memory Elliptic Curve Cryptography library from scratch, optimized for resource-constrained systems.