Skip to content
View Khushi-Dua's full-sized avatar
  • Sharda University
  • Delhi, India

Block or report Khushi-Dua

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Khushi-Dua/README.md

Hi πŸ‘‹, I'm Khushi Dua

Building AI solutions for real-world impact πŸš€

BTech CSE (AI & ML) | Aspiring Data Scientist | Machine Learning & Deep Learning


About Me

I am a Data Scientist in training specializing in time-series forecasting, anomaly detection, and deep learning systems. I build end-to-end machine learning pipelines focused on predictive accuracy, optimization, and real-world impact.

πŸ”Ή Experienced with large-scale datasets (12M+ records)
πŸ”Ή Focused on sustainability analytics & intelligent prediction systems
πŸ”Ή Passionate about solving real-world problems using AI-driven solutions

πŸŽ“ BTech CSE – Artificial Intelligence & Machine Learning
🏫 Sharda University | Expected Graduation: June 2027


Technical Skills

Programming & Libraries

Python SQL Pandas NumPy Scikit-learn TensorFlow Keras

Machine Learning

Time Series Forecasting LSTM GRU ARIMA Regression Classification Clustering Feature Engineering Hyperparameter Tuning Model Evaluation Anomaly Detection

Data Science & Tools

EDA Statistical Analysis Data Wrangling Power BI Excel Git GitHub Jupyter Notebook Google Colab Kaggle


Featured Projects

⚑ Electricity Consumption & Anomaly Detection

  • Built hybrid LSTM-GRU architecture on 12M+ records.
  • Achieved 70% RMSE reduction over traditional models.
  • Implemented residual-based anomaly detection with 99.94% accuracy.

🌱 Green Core – Carbon Intelligence Dashboard

  • Developed carbon-aware forecasting models.
  • Enabled sustainability-driven deployment insights.
  • Achieved 85%+ prediction accuracy.

🧠 Brain Tumor Detection Using Deep Learning

  • Designed CNN model for MRI classification.
  • Achieved 95.18% accuracy and improved generalization.

πŸ’‘ All projects include complete data preprocessing, model evaluation, and performance comparison workflows.


πŸͺΆ Open Source Badges

GSSOC(24) Badges πŸͺΆ

Open Source Contributions

πŸ”Ή Social Winter of Code (SWOC) – Season 5

  • Automated preprocessing pipelines reducing manual effort by 60%.

πŸ”Ή GirlScript Summer of Code (GSSOC) '24 Extended

  • Ranked 641 among 60,000+ contributors.
  • Improved model performance through preprocessing optimization.

GitHub Stats


πŸ“œ Certifications

image

  • HackerRank SQL (Basic) β€” Verified
  • HackerRank SQL (Intermediate) β€” Verified
  • HackerRank SQL (Advanced) β€” Verified

Currently Learning

  • Advanced SQL for Data Analytics
  • Data Structures & Algorithms
  • Real-time ML Pipelines
  • Model Deployment & MLOps Basics

Currently focusing on improving model deployment and production-ready ML workflows.

Connect With Me


⭐ Open to Data Science & Machine Learning Internship Opportunities.

Education

  • B.Tech in Computer Science & Engineering (AIML), Sharda University β€” 2023-Present
  • Higher Secondary School (12th) ,Vishal Bharti Senior Secondary Schoolβ€” [2022-2023]

Goals

  • Build production-ready machine learning systems and deploy scalable AI solutions.
  • Contribute to impactful open-source AI and data science projects.
  • Apply machine learning to sustainability, healthcare, and intelligent analytics systems.

Pinned Loading

  1. Khushi-Dua Khushi-Dua Public

    Readme. md

    1

  2. BOLT BOLT Public

    BOLT is an AI-based electricity consumption forecasting and anomaly detection system using ARIMA, Random Forest, GRU,LSTM and Hybrid LSTM_GRU models to predict energy usage patterns and identify ab…

    Jupyter Notebook

  3. Brain-Tumor-Detection Brain-Tumor-Detection Public

    A deep learning-based brain tumor detection system using a Convolutional Neural Network (CNN) to classify MRI images into tumor and non-tumor categories. The model is trained on 7,000+ brain MRI sc…

    Jupyter Notebook

  4. Machine-Learning Machine-Learning Public

    Welcome to my Machine Learning repo! πŸŽ‰ I'm a beginner documenting my journey as I explore datasets and experiment with NumPy, Pandas, and machine learning algorithms. Feel free to dive in, check ou…

    Jupyter Notebook 1

  5. Data-Science Data-Science Public

    This repository serves as a centralized platform for our teacher to post classwork completed during lessons and assign Continuous Assessments (CA) based on that work. It helps maintain an organized…

    Jupyter Notebook

  6. Voice-Activated-Personal-Assistant Voice-Activated-Personal-Assistant Public

    Your Intelligent, Voice-Powered Helper

    Python