Skip to content
View rohithdonthi's full-sized avatar

Highlights

  • Pro

Block or report rohithdonthi

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.

Content in all repositories owned by your account will be closed.
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
RohithDonthi/README.md

Hi, I'm Rohith 👋

Data Scientist specializing in credit risk modeling and energy market analytics, with an MS in Data Science (Statistics Track) from Rutgers University.

I build end-to-end ML pipelines - data → modeling → calibration → monitoring - and I'm especially interested in problems where statistical rigor and domain context both matter: credit underwriting, energy markets, and policy analysis.

Featured projects

Credit Underwriting & Decisioning System Production-style underwriting pipeline: calibrated PD model → scorecard (PDO) mapping → decision/reason codes → PSI drift monitoring, built on a fully synthetic, compliance-documented credit bureau dataset.

Energy Market Forecasting & Policy Impact Analysis Forecasts U.S. state-level electricity generation mix (50 states, 1990–2023) using CLR-transformed compositional time series and per-state VAR models - including an honest benchmark against a persistence baseline.

Background

  • Data Science Intern, Synergy Resources - built a credit underwriting pipeline (AUC 0.87) supporting a 50K+ account portfolio
  • Statistics Grader & Mentor, Rutgers University (2 years)
  • Certifications: Bloomberg Market Concepts (BMC), Financial Statistics & Risk Management (FSRM)

Toolbox

Python XGBoost LightGBM TensorFlow scikit-learn statsmodels pandas · AWS (S3, EC2, Lambda) Docker · SQL

Let's connect

LinkedIn

Pinned Loading

  1. credit-underwriting-decisioning-system credit-underwriting-decisioning-system Public

    End-to-end underwriting decisioning pipeline with time-based validation, calibrated probability outputs, scorecard mapping, and monitoring-ready artifacts.

    Jupyter Notebook

  2. Energy-Market-Forecasting-and-Policy-Impact-Analysis Energy-Market-Forecasting-and-Policy-Impact-Analysis Public

    Forecasting U.S. state-level electricity generation mix (50 states, 1990–2023) using CLR transformation and VAR models - benchmarked against a persistence baseline.

    Jupyter Notebook

  3. Online-News-Popularity-Prediction Online-News-Popularity-Prediction Public

    Predicting online news article popularity using regression, regularization, and ANN models in R and Python.

    Jupyter Notebook

  4. rohithdonthi.github.io rohithdonthi.github.io Public

    HTML