Fraud Detection System
Built a machine learning model to detect fraudulent transactions in financial datasets with high precision and recall. The system uses both supervised and ensemble learning techniques to achieve robustness.
Tools Used: Python, Scikit-learn, XGBoost, Pandas
Speech-to-Text System
Created a deep learning pipeline to convert speech into text using attention-based models, ensuring high accuracy for real-world audio data.
Tools Used: PyTorch, Librosa, TensorFlow
Document Classification
Implemented an NLP pipeline to classify legal documents into predefined categories using fine-tuned BERT embeddings and transfer learning.
Tools Used: Python, Hugging Face Transformers, BERT, NLTK
Predictive Maintenance
Designed a time-series forecasting model to predict equipment failures and optimize maintenance schedules for industrial processes.
Tools Used: Python, TensorFlow, Keras, Pandas
Inventory Management System
Developed a full-stack inventory management application to track stock levels and automate procurement processes.
Tools Used: Angular, Spring Boot, PostgreSQL
Cloud File Storage App
Designed a secure file storage and sharing platform integrated with AWS S3, enabling seamless cross-device file management.
Tools Used: React, AWS S3, Node.js
Real-Time Collaboration Tool
Built a web application for real-time collaboration on documents, featuring live edits and user tracking.
Tools Used: WebSockets, Node.js, React