Skip to content
View aksharabharath's full-sized avatar

Block or report aksharabharath

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 is supported. This note will only be visible to you.
Report abuse

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

Report abuse
aksharabharath/README.md

Hi, I'm Akshara! 👋

Machine Learning • Computational Science • AI Systems

I'm a sophomore at American High School exploring how machine learning can be used to study scientific problems and build real-world AI systems.

My projects span astrophysics, space weather, remote sensing, time-series forecasting, and AI-powered automation. I enjoy working with real datasets, building end-to-end ML pipelines, and going beyond model accuracy to investigate interpretability, robustness, uncertainty, and the scientific meaning behind predictions.


What I Build

Computational Science & Scientific ML

I use machine learning to investigate problems in astronomy, space science, and Earth observation.

stellar-scope Exploring stellar and exoplanetary data through computational analysis and machine learning.

geomagnetic-storm-forecasting-ml A physics-informed LSTM time-series model for forecasting geomagnetic storms from solar-wind data. The project combines temporal deep learning with space-weather feature engineering.

exoplanet-classification-ml A scientific ML study investigating whether host-star properties can predict gas-giant classification in observed exoplanets, with interpretability, uncertainty, and robustness analysis.

satellite-landcover-knn A K-Nearest Neighbors model for classifying EuroSAT satellite imagery into land-cover categories, exploring classical ML for remote sensing and computer vision.


Applied AI & Automation

I also build AI systems that connect machine-learning models and language models to practical workflows.

vapi-appointment-reminders A voice-AI automation project using Vapi to build automated appointment reminder and calling workflows.

I'm interested in the engineering side of AI as well—from model and API integration to automation, data pipelines, and usable interfaces.


My Approach

I like projects that go beyond simply training a model.

My work often involves:

  • Real-world scientific datasets
  • Data cleaning and preprocessing
  • Feature engineering based on domain knowledge
  • Classical and deep learning approaches
  • Time-series modeling
  • Model evaluation and robustness testing
  • Explainable ML and SHAP
  • Uncertainty analysis
  • Data visualization
  • Reproducible computational workflows

I'm particularly interested in the question:

Can machine learning help us understand something about the real world—not just predict it?


Technical Stack

Area Technologies
Languages Python, JavaScript, HTML/CSS
Machine Learning Scikit-learn, XGBoost, LSTM, KNN, Random Forest
Scientific Computing NumPy, Pandas, SciPy
ML Analysis Feature Engineering, SHAP, Model Evaluation, Uncertainty & Robustness Analysis
Visualization Matplotlib, Seaborn, Plotly
AI / Automation Vapi, LLM APIs, Voice AI
Applications Streamlit, React, Firebase
Tools Git, GitHub, Jupyter, VS Code, Google Colab

Areas of Interest

  • Computational Astrophysics
  • Exoplanet & Stellar Science
  • Space Weather
  • Geomagnetic Storm Forecasting
  • Remote Sensing & Earth Observation
  • Scientific Machine Learning
  • Time-Series Modeling
  • Explainable & Physics-Informed ML
  • AI Systems & Automation

Featured Projects

Project Focus
stellar-scope Stellar & exoplanet computational analysis
geomagnetic-storm-forecasting-ml Physics-informed LSTM space-weather forecasting
exoplanet-classification-ml Scientific ML + astrophysics + interpretability
satellite-landcover-knn KNN + satellite imagery + remote sensing
vapi-appointment-reminders Voice AI + workflow automation

Let's Connect

I'm interested in connecting with students, researchers, and developers working in machine learning, computational science, astronomy, space science, and AI systems.

🔗 LinkedIn

Explore my repositories to see what I'm building at the intersection of science, machine learning, and AI engineering.

Pinned Loading

  1. vapi-appointment-reminders vapi-appointment-reminders Public

    TypeScript

  2. satellite-landcover-knn satellite-landcover-knn Public

    Machine learning project using K-Nearest Neighbors (KNN) to classify EuroSAT satellite images into Forest, River, and Residential land cover classes.

    Jupyter Notebook

  3. exoplanet-classification-ml exoplanet-classification-ml Public

    Scientific machine learning study investigating whether host-star properties can predict gas-giant classification in observed exoplanets, with interpretability, uncertainty, and robustness analysis.

    Jupyter Notebook

  4. womens-50free-varsity-cutoff-analysis womens-50free-varsity-cutoff-analysis Public

    Analyzing varsity cutoff trends for the 50-yard freestyle in high school girls’ swimming using real performance data. Includes statistical insights, time ranges, and data visuals.

    Python