FinSights
An AI-powered financial analysis platform that integrates large language models (LLMs), real-time data APIs, and multimodal document parsing.
FinSights automates the ingestion, retrieval, and interpretation of structured and unstructured financial data across formats such as PDFs, spreadsheets, and earnings call transcripts.
Child Mind Institute - Problematic Internet Use
A Kaggle Competition with the goal to develop a predictive model that analyzes children's physical activity and fitness data to identify early signs of problematic internet use. Identifying these patterns can help trigger interventions to encourage healthier digital habits.
Tennis in the near future: using computer vision for automatic performance analysis
Assisted PHD project applying computer vision and movement analysis algorithms on tennis games.
Sentiment Classification with Pretrained Embeddings
Performed sentiment classification on the Rotten Tomatoes dataset using various deep learning models built on top of pretrained word embeddings. Explored different model architectures and techniques to handle out-of-vocabulary (OOV) words.
PostgreSQL Query Plan Visualize
A React based web application, a tool for PostgreSQL query visualization using Python Flask framework to understand the SQL query execution plan, memory utlized and dissect access by sub-trees. It is designed with a custom sub querying backend algorithm to identify bottlenecks in the execution plan.
Database Design - NBA Games Dataset
Designed and implemented an optimized database for a large NBA games dataset in C++, utilizing B+ Tree indexing and unspanned storage system for reduced disk I/O and improved performance.