Mobile App Developer & Researcher
I am a researcher and mobile app developer specializing in on-device AI, computer vision, and applied machine learning, with additional research in legal information retrieval and deepfake detection.
Currently contributing to Siamese deep metric learning for non-invasive biometric livestock identification, in collaboration with the United Nations Development Programme (UNDP) in Nepal.
Project
Muzzle-Based Livestock Identification via Siamese Deep Metric Learning
Collaboration
A joint initiative between Plantsat and the United Nations Development Programme (UNDP) in Nepal, working toward non-invasive biometric identification of individual livestock animals using deep metric learning and computer vision.
What I am working on
Multi-backbone training pipeline implemented and evaluated. Grad-CAM interpretability validated. Multi-environment and post-mortem dataset collection ongoing to improve model robustness and accuracy.
Published work in Nepali legal information retrieval, embedding model evaluation, and deepfake & image forgery detection.
An empirical study of embedding choices for retrieving Nepali legal documents in a low-resource setting.
Study contribution: Compares retrieval behavior rather than treating embedding choice as an implementation detail, with attention to the constraints of Nepali legal text.
A comparative study of Convolutional Neural Networks and Transformer models on a custom-curated dataset for deepfake detection.
Study contribution: Created a custom-curated dataset and evaluated CNN versus Transformer architecture performance on image forgery and deepfake detection.
A retrieval-augmented legal analysis pipeline designed for access to information in the Nepalese legal context.
Study contribution: Processes legal documents for retrieval, expands user queries, and evaluates rule-recall, rhetorical-understanding, and interpretation-oriented questions.
Selected mobile and systems software projects — cross-platform applications, real-time communication architectures, on-device compute, and collaborative research tooling.
An integrated healthcare management application featuring real-time appointment booking, consultation chat, video consulting, and a built-in medical products e-commerce system.
Systems focus: Real-time communication protocols (WebRTC), secure chat and video streaming architectures, and e-commerce transaction handling.
A collaborative Android application developed with students from MIT and NYU that detects and measures light pollution from lunar photographs. The app handles camera-based image capture, analysis, and all computation entirely on-device.
Systems focus: On-device image processing, camera frame analysis, and performance-optimized computation without network dependency.