Innovative solutions built with cutting-edge technologies
I built LexiMind, a multi-task NLP system based on FLAN-T5 that jointly performs abstractive summarization, topic classification, and emotion detection on literary and academic text. The project focused on understanding when multi-task learning helps versus hurts performance, rather than assuming joint training is always beneficial.
End-to-end machine learning system forecasting Formula 1 race finishes and qualifying performance using the FastF1 datset.
Content-based filtering using TF‑IDF and cosine similarity to generate personalized recommendations over 10k+ movies.
Engineered a comprehensive multi-sport and event web application using modern web technologies. Integrated various APIs for real-time sports data, event tickets, and schedules.
Developed a personal portfolio website using HTML, CSS, and JavaScript to showcase projects and skills. Utilized Netlify for free hosting and GitHub for version control.
Tableau dashboard analyzing Formula 1 data from 1950–2024. Visualizes drivers, constructors, and circuits across F1 history to uncover performance trends and insights.
Developed a Raspberry Pi-based monitoring system for fossil fuel power plants, measuring temperature variations and structural changes with cloud integration.
Data-driven evaluation of Intel's 2024 device repurposing and recycling program to prioritize devices for maximum reductions in e-waste, energy consumption, and CO₂ emissions.
Conducted comprehensive data analysis for The Recording Academy's websites, providing data-driven insights into audience engagement and business metrics.