Projects
A collection of cutting-edge projects built to solve real-world efficiency problems — spanning AI/ML, cloud engineering, LLM apps, intelligent automation, and impactful research platforms.
ADRPredictor: Adverse Drug Reaction Risk Analyzer
🧠 ADRPredictor uses real-world FAERS data and machine learning to predict serious adverse drug reaction (ADR) risks. Includes local LLM-powered explanations via Ollama.
DiagnoAid: AI-Powered Disease Predictor
🩺 DiagnoAid is an intelligent medical assistant that uses dual ML models and symptom severity weighting to predict possible diseases. Offers LLM-based explanations and safety tips via Ollama.
DrugCompare: Intelligent Drug Comparison Tool
💊 DrugCompare allows users to select two approved medications and instantly compare their medical details – including indications, absorption, toxicity, and more – with AI-powered summaries via Ollama.
TrialMatchAI: Clinical Trial Eligibility Matcher
🧪 TrialMatchAI intelligently matches patients with clinical trials using FAISS-based semantic search and LLM-powered eligibility evaluation. It analyzes trial criteria, user conditions, and outputs match scores with reasoning.
Gemini Quizify: AI‑Powered Quiz Builder
📝 Gemini Quizify generates custom multiple‑choice quizzes from any document or topic using Google Gemini & Vertex AI embeddings. Built with LangChain and Streamlit, it offers interactive quiz navigation and answer explanations.
Movie Recommendation System
🎬 Personalized movie recommender using collaborative and content-based filtering with user feedback loop. Built with Python, Pandas, scikit-learn, and Streamlit.
Financial Dashboard
📊 Interactive financial dashboard providing stock & crypto tracking, portfolio analytics and news aggregation. Built with Streamlit and Plotly.
AI Chatbot
🤖 AI-powered conversational agent with intent recognition and context persistence. Built with NLP, Rasa, or similar frameworks and Python backend.
Housing Price Predictor
🏠 Housing price prediction model using regression algorithms on real estate datasets. Includes evaluation metrics and feature analysis.
Diabetes Onset Predictor
💡 Predicts diabetes onset using clinical indicators and multiple ML models. Includes evaluation and ROC analysis for model performance.
Video Object Filter Detection App
🎥 Video stream object detection and filtering application using OpenCV and TensorFlow. Includes real-time filters and detections.
ML Feature Comparison Tool
⚙️ Compares feature importance across multiple ML models with visual metrics and sortable views. Built with Python and Streamlit.
MNIST Deep Vision
📷 Deep neural network trained on MNIST for robust digit classification, with accuracy visualization and model tuning.
MNIST CNN Digit Recognition
✏️ Convolutional Neural Network for handwritten digit recognition on MNIST, achieving high accuracy with data augmentation.