I'm a passionate Data Science, ML and DL Learner currently building my foundation in Python, NumPy, Pandas, EDA, and SQL. My goal is to grow into a skilled AI and Machine Learning engineer, working on projects that turn data into actionable insights.

• 🌱 Currently learning Machine Learning, Data Visualization, and advanced AI concepts.
• 💬 Ask me about Python, Data Analysis, SQL, or beginner-friendly ML concepts.
• ⚡ Fun fact: I enjoy exploring datasets and discovering hidden patterns!

I am currently working on beginner-friendly Data Science, Machine Learning and Deep Learning Projects to apply my skills:

A PyTorch-powered handwritten digit recognizer that lets users draw or upload digit images (0–9) and predicts them using a convolutional neural network (CNN) via a simple Streamlit interface. Great for learning, digitization tasks, and lightweight AI demos.

• Key Features: Handwritten digit drawing canvas, image upload prediction, pre-trained CNN model, fast inference with high accuracy, modular Streamlit apps.
• Tools: Python, PyTorch, Torchvision, Streamlit, NumPy, PIL.

An end-to-end regression solution designed to monitor and predict employee burnout levels (0-100) based on work habits and lifestyle metrics. Features a sophisticated Glassmorphic UI and a robust pipeline that handles everything from data ingestion to real-time risk assessment.

• Key Features: Proactive risk alerts (Low/Medium/High), automated data transformation, and dual-deployment (Flask & Streamlit).
• Tools: Python, Scikit-learn, XGBoost, CatBoost, Flask, Streamlit, Pandas.

Predicts students’ math scores using demographic and academic attributes. This project implements a complete industry-standard ML lifecycle, including comprehensive EDA, feature engineering, and model serialization.

• Key Features: Interactive data visualization, hyperparameter tuning, and a user-friendly API for instant predictions.
• Tools: Pandas, Scikit-learn, CatBoost, XGBoost, Flask, Streamlit, Altair.

An interactive OCR and translation web application for extracting text from images using region-based selection and full-image scanning. Built as a lightweight computer vision application with a clean, production-ready UI powered by Streamlit and deep-learning OCR via EasyOCR, with on-the-fly translation backed by Google.

• Key Features: Google-Lens-style crop selection, full image scan mode, multi-language OCR (English, Nepali, Hindi), instant translation, one-click copy buttons, and downloadable session history (CSV).
• Tools: Python, Streamlit, EasyOCR, streamlit-cropper, deep-translator, Pillow, NumPy, Pandas.

Cleans and analyzes scraped job market data and produces visual insights using exploratory data analysis. Tools: Pandas, Matplotlib, Seaborn

Automated web scraping pipeline that collects job listings, removes duplicates, sorts records, and exports results to CSV with scheduled execution. Tools: Python, Selenium, GitHub Actions

CLI tool that generates project folder structures from tree-style text files and is published on PyPI for fast and consistent project setup. Tools: Python, Click, setuptools

Command-line tool to extract and export YouTube playlist metadata for further processing. Tools: Python, yt-dlp

More projects will be added as I progress on my Data Science and ML journey.

“Learning today, building AI for tomorrow.” 🚀