- 🔭 A software developer at heart
- 🌱 Passionate about Mobile Development, Artificial Intelligence, Augmented Reality (AR), and Virtual Reality (VR)
- 🤖 Exploring and collaborating on AR/VR applications
- 👨💻 Curious about new technologies & adaptability believer
- ⚡ Fun fact: Die-hard fan of Marvel Studios and Anime
Jetpack Compose · Room · Dagger · RxJava · Firebase · JUnit/Mockito/Espresso · Picasso/Glide · Retrofit
A privacy-focused productivity app that transforms your notes into an intelligent workspace. Powered natively by on-device Gemma models, Gemini Nano, and Gemini 1.5 Flash, it provides cutting-edge agentic AI features—including a full conversational AI Assistant with tool-calling capabilities that organize, summarize, and act on your content securely and autonomously!
🛠️ Built with: Kotlin · Jetpack Compose · Dagger-Hilt · Room · Supabase
A modern Android application designed for food enthusiasts to discover, track, and navigate to restaurants they want to visit. Engineered using an offline-first MVVM architecture powered by local caching, it connects seamlessly to the Google Places API to provide interactive maps, personalized wishlists, and seamless widget tracking!
🛠️ Built with: Kotlin · Jetpack Compose · Dagger-Hilt · Retrofit · Room
A weather forecasting app that provides real-time local weather updates, including hourly temperature, wind speed, atmospheric pressure, humidity, UV index, and visibility distance. It also offers radar maps, air quality data, outdoor sports index, sunrise and sunset times, and more. This app ensures you travel with confidence by delivering accurate weather information for the next 24 hours and beyond.
🛠️ Built with: Kotlin · Jetpack Compose · Firebase
Interactive Quiz App for immersive learning analytics. Designed to engage users while tracking performance and progress.
🛠️ Built with: Python · Streamlit · Pandas · Plotly
An augmented reality (AR) application that brings the iconic ForeverSpin spinning top into the real world. Utilising ARKit, this project enables users to place and interact with a 3D spinning top in their environment, providing an immersive and engaging experience.
🛠️ Built with: Unity · C#
The official code for the paper "DS6: Deformation-aware learning for small vessel segmentation with small, imperfectly labelled dataset". This project addresses challenges in medical image segmentation by proposing a deformation-aware learning approach, tailored for small vessels and datasets with limited annotations, enabling more accurate vessel detection in medical imaging.
🛠️ Built with: Python · Pytorch · Pandas · Numpy
Part of the master thesis titled "Synth2Real: 3D-Furniture Reconstruction in Ersatz Environment (S2R:3D-FREE)", this Unity-based pipeline creates synthetic datasets for 3D reconstruction. It imports rooms from SceneNet and furniture from Pix3D, allowing users to select categories of furniture and generate images with specified camera angles. This project is coupled with a deep learning pipeline available at 3DReconstruction.
🛠️ Built with: Unity · C#
A deep learning pipeline for reconstructing 3D objects from multi-view images, often used alongside synthetic datasets generated by 3dScene. It leverages PyTorch and OpenCV to produce accurate 3D models from 2D images, supporting research in 3D vision and computer graphics.
🛠️ Built with: Python · Pytorch · Pandas · Numpy

