🚀 About Me               I'm a passionate Data Enthusiast and Open Source Contributor from India, with a keen interest in leveraging data to build intelligent solutions. I thrive on exploring new datasets, uncovering hidden patterns, and collaborating on impactful projects.                     🔭 I’m currently working on Data Analysis and Machine Learning Projects.         🌱 I’m mastering Advanced ML, Deep Learning, and Data Engineering concepts.         👯 I’m looking to collaborate on innovative open-source data science projects.         💬 Ask me about anything related to Python, SQL, Statistics, or Data Visualization.         📫 How to reach me: [email protected]         ⚡ Fun fact: I believe every dataset has a story to tell!

📊 GitHub Stats

    I believe in the power of community and collaborative learning.

I am a dedicated contributor to ArviZ, the essential Python library for Bayesian model analysis (diagnostics, comparison, and posterior predictive checks). My work focuses on building robust statistical functions and high-utility visualization tools to advance Bayesian data science.

🔗 Check out my in-depth project overview: ArviZ Plotting Refactoring Initiative Loom Video

• Model Comparison Tooling (arviz-stats)

  • Bayes Factor Implementation (PR #52 & #104): Developed the core bayes_factor() function to enable rigorous statistical model comparison in Bayesian frameworks, including support for multi-variable input and customizable prior odds.
  • KDE & Bayes Factor Refactor (PR #95): Improved the internal logic for Kernel Density Estimation (KDE) and Bayes Factor computations for better accuracy, readability, and performance.
  • 
  • 

• High-Utility Visualization (arviz-plots)

  • Plotting Predictive Intervals (PR #334): Developed plot_ppc_intervals() for Bayesian Model Calibration, a critical visualization that plots Posterior Predictive Intervals against observed data to diagnose model fit across all backends (Matplotlib, Bokeh, Plotly).
  • Bayes Factor Visualization (PR #158): Implemented plot_bf() to visually represent Bayes Factor outputs, significantly improving user interpretation of model comparison results.
  • Rug Plot Support (PR #192): Added rug_plot functionality to plot_dist(), enhancing insight into 1D marginal posterior distributions.
  • 
  • 

• Documentation & Testing

  • Onboarding Documentation (PR #2444): Updated the official “Getting Started” guide in the legacy ArviZ repo to reduce friction and improve clarity for new users.
  • Test Enhancements (PR #177): Extended internal data generation and test reliability for multiple statistical plotting functions in arviz-plots.
  • 

I share my knowledge and experience through blog posts explaining foundational Bayesian concepts and the intuition behind my open-source work.

Blog Post Title	Key Topics Covered
My Journey with Bayes Factor (Part-1)	Introduces Bayes’ theorem using a real-world example, explains Bayesian models, and covers how ArviZ helps visualize prior/posterior distributions.
My Journey Contributing to ArviZ (Part-2)	Details my contributions: the bayes_factor() function (Savage-Dickey method) and the plot_bf() visualization, showing how these tools aid model evaluation.
Read them on Medium: