The whole reason I build things is not just to learn new technologies, but to help people and make their lives easier through technology. A few projects that reflect that goal:
Led a multidisciplinary team to design and implement a system that measures cognitive load using EEG signals, facial emotion recognition, gameplay, and NASA-TLX surveys. I owned full-stack integrations, real-time data pipelines, and system performance, improving responsiveness and measurement precision.
Built an end-to-end workflow that uses LLMs to analyze unstructured broker reports and convert them into structured investment signals. The pipeline scales across thousands of documents and significantly improved automation compared to a previous NLP approach.
An AR-powered wardrobe assistant that uses the OpenAI API and React to generate personalized outfit recommendations for young professionals. Our goal was simple: help people feel confident before they step out the door.
Evaluated multiple machine learning models for detecting credit card fraud, including logistic regression and neural networks. Tuned models with metrics like precision, recall, and F1-score to reduce false positives while keeping high accuracy.