Python

What I Built Maintenance-Eye is a real-time multimodal maintenance copilot for field technicians. A technician can point a phone camera at equipment, speak naturally, and get back live visual analysis, voice responses, equipment lookup, maintenance context, and work-order support. This project was built for the Google Gemini Live Agent Challenge and is deployed as a working web application. Why This Workflow Matters Maintenance work happens in noisy, high-stakes environments where technicians are moving, inspecting, and handling tools. Traditional maintenance software assumes the user can stop, type, search, and document everything manually. ...

Overview An earlier machine learning deployment project for house price prediction, built as a Flask application with an AWS deployment workflow and automated GitHub Actions pipeline. Architecture Flask application — serves the prediction interface and inference flow AWS ECR — stores the Docker image built by the CI pipeline AWS EC2 — hosts the deployed container and self-hosted GitHub runner GitHub Actions — runs CI and pushes the deployment image Features Interactive prediction UI — users can change housing features and see updated predictions Containerized deployment workflow — Docker image built and pushed through GitHub Actions End-to-end project structure — data ingestion, transformation, training, and prediction pipelines in one codebase Tech Stack Python (Scikit-learn, Pandas, Flask) AWS (ECR, EC2) GitHub Actions Docker Links GitHub Repository Key Takeaway This project was an early end-to-end ML deployment exercise focused on packaging a prediction model as a containerized web app with an AWS deployment workflow and CI/CD automation. ...

Overview Designed classification models for Autism Spectrum Disorder (ASD) detection using Support Vector Machines and Convolutional Neural Networks, achieving 94% accuracy. This research was published in the International Journal of Information Technology and Computer Science (IJITCS). Publication Published in: International Journal of Information Technology and Computer Science (IJITCS), Vol.14, No.4 Approach Data preprocessing — Cleaned and prepared ASD screening datasets with feature engineering Model development — Implemented SVM and CNN architectures for binary classification Evaluation — Compared model performance using accuracy, precision, recall, and F1-score Visualization — Created Tableau dashboards to help healthcare professionals understand demographic patterns Tech Stack Python (Scikit-learn, TensorFlow) Support Vector Machines Convolutional Neural Networks Tableau Results 94% accuracy on ASD detection CNN outperformed traditional ML approaches on this dataset Visualization dashboards provided actionable insights for healthcare professionals Key Takeaway This project combined rigorous ML methodology with healthcare domain application, demonstrating how deep learning can assist in early screening for developmental conditions. ...