My Projects

This is an ever-evolving list, but might not be entirely up to date. For more of my projects, you may visit my Github Profile. I'm a curious sort, so if you have any questions or suggestions about my work, feel free to share!

  • Space Engineering

    Analysis of Atmospheric Drag on VLEO Satellites with Variable Solar Array Configurations

    Modelled atmospheric drag for the satellites operating in Very Low Earth Orbit (VLEO) using the MSIS-00 atmospheric model and CelestLab simulations. Satellite dynamics were analyzed under varying solar activity conditions and over various solar array configurations [face-on, edge-on, and sun-tracking] to quantify their impact on orbital decay.

    Talk Abstract [YSC 2026, Russia] Published Case Study Download Report Presentation Slides
    Web Design

  • AI for Sustainability

    Facilitating citizen science initiatives for waste management through Machine Learning

    This project utilizes Scilab as a free, open-source platform for sustainable waste management. It implements two studies: neural network-based image classification system to automatically sort waste into four categories [Hazardous, Non-Recyclable, Organic & Recyclable], and K-means clustering to identify different air quality zones.

    Poster [International Citizen Science Conclave, IISER Pune] Published Case Study Download Report
    Web Design

  • Applied Optimization

    Optimization in Scilab using the gradient descent algorithms

    This case study secured third position in the global Scilab Hackathon 2025. It provides practical recommendations for choosing the right gradient descent strategy depending on the task requirements. It compares three optimization algorithms: Batch (BGD), Stochastic (SGD), and Mini-Batch (Mini-BGD), applied to logistic regression for binary classification.

    View Project Hackathon Results Download Report
    Web Design

  • Machine Learning in Python

    System Threat Detection Using Ensemble Learning

    Developed a binary classification model to predict system security threats by training and comparing Logistic Regression, Random Forest, and XGBoost on a noisy, real-world dataset. The final optimized XGBoost model was selected for test set predictions, demonstrating an end-to-end ML workflow from EDA to deployment-ready inference.

    View Project
    Web Design

  • Computational Physics

    Identification of elements and analysis of their properties

    This project aims to identify unknown or newly synthesized elements by computing their atomic properties based on the Bohr model. From the experimental atomic mass, key physical quantities are calculated and benchmarked against theoretical values, verifying the Ritz combination principle and estimating the atomic number.

    Published Case Study Download Report
    Web Design

  • Flight Dynamics

    Simulation of pitch of an aircraft

    This project simulates the pitch control of an aircraft using a PID (Proportional - Integral - Derivative) controller built in Xcos. The longitudinal dynamics of the aircraft are modelled as a transfer function, and closed-loop system is simulated with varying feedback gain values to determine the best-stabilized pitch response.

    Published Case Study Download Report
    pid
envelop icon

Feel free to reach out!

I'm open to research collaborations, academic discussions, and volunteering for causes that drive social good.