
Audocare AI Voice Agent
Intelligent, speech-driven assistant that engages users in real-time conversations to detect and assess symptoms of illness

Intelligent, speech-driven assistant that engages users in real-time conversations to detect and assess symptoms of illness

A companion frontend for the Audocare AI Voice Agent, built using Next.js and LiveKit SDK.

Build and deploy RAG-based AI applications using LangChain and IBM tools.

Projects and coursework for the Google Advanced Data Analytics Certification.

Flask-based backend services integrated with AI capabilities on IBM Cloud.

Detecting concrete cracks with high accuracy using transfer learning with ResNet50 and VGG16.

A CNN model built with PyTorch to classify Fashion-MNIST images with over 85% accuracy.

A logistic regression model in PyTorch to predict match outcomes based on in-game stats.

Image classifier for waste sorting using VGG16 and transfer learning techniques.

A project exploring polynomial regression techniques in predictive modeling.

Basic implementation and visualization of linear regression models.

My personal portfolio built with Next.js and TypeScript.