Projects
Smart Write: A Context-based Nepali Text Autocompletion
2022 - 2023Undergraduate research - Under the guidance of Mr. Bal Krishna Nyaupane(Associate Professor, IOE, TU)
- Evaluated the performance of transformer and sequence-to-sequence models (LSTM and GRU) for sentence autocompletion in the Nepali language.
- Implemented a bidirectional LSTM model with attention mechanism for context-aware text generation.
- Trained a Word2Vec model for the Nepali language
BeeWell
Jan 2023 Deltathon, Delta 3.0, Purwanchal Campus, Tribhuvan University.
Built an AI-based system for detecting bees, monitoring their behavior and health status, and suggesting possible solutions to the detected problems directly to farmers.
Won the Best Hardware Project Award in the hackathon.
Retinal Blood Vessels Segmentation
Oct 2025
Semantic Segmentation of Retinal Blood Vessel Images using U-Net Model.
Trained different configurations of U-Net models (with 2, 3 and 4 blocks) with highest dice score of 0.8 .
Enzyme Molecules Classification Using Graph Neural Networks (GNNs)
Nov 2025Classification of different types of enzyme molecules using Graph Convolution Networks (GCN), Graph Attention Networks (GAT)
Heartbeat Classification
Oct 2025Classification of different types of heartbeat based on Electrocardiogram (ECG) signals record using Convolutional Neural Networks (CNN) , identifying possible abnormal heart rhythms.
Molecule Solubility Prediction Using Graph Convolutional Network
Sept 2025Modeled molecules using Graph Neural Network (GNNs) to predict their solubility.
Malaria Cell Images Classification
May 2023Classfication of malaria cell images to identify infected malaria cell.
Easy Blood System
2022A dedicated platform for patients to get help when there is need for blood and where donors can post their availability of blood.
Personal Stock Portfolio Tracker
2022An application for investors where they can maintain their stock portfolio, receive notifications on the profit and loss amounts, and discuss various queries about the stock market using the discussion forum.