PhD in Computer Science (Machine Learning & Responsible AI, 2025) | Independent Director (IICA, Govt. of India) | Open Science Advocate

I specialize in fairness-aware, interpretable AI models for credit scoring, carbon scoring, and financial technology applications.
My research integrates alternative datasets (psychometric, digital, utility, rural carbon data) to build Responsible AI systems aligned with global ESG and sustainability goals.

I curate the DeepaShukla Dataverse – Digital Credit Scoring (Harvard Dataverse) and maintain multiple GitHub repositories to ensure open, reproducible research.

• 🔭 I’m currently working on: Farm-to-Finance Responsible AI models for carbon & credit scoring
• 🌱 I’m currently practicing: GenAI pipelines, ESG frameworks, and Responsible AI governance
• 👯 I’m working on: Open datasets, fairness-aware ML, and sustainability-focused AI projects
• 🤔 I’m looking for help with: Scaling dual-task AI models into production-ready ecosystems
• 💬 Ask me about: Credit scoring, carbon scoring, fairness in ML, open science, ESG-aligned Responsible AI
• 📫 How to reach me: LinkedIn | Google Scholar | ORCID | Harvard Dataverse
• 😄 Pronouns: She/Her
• ⚡ Fun fact: I bridge AI, governance, and sustainability — and also once built an Arduino-based plant watering system together with my son Chanakya Shukla, a qualified engineer specializing in water governance. 🌱💧

• ML-Credit-Scoring-ThinFile-Consumers → Final implementation repo for my PhD thesis (with GitHub Pages).
• PhDP → Python codes for all tables in PhD Thesis, Chapter 5 (Results & Evaluation).
• hypothetical_model → Python implementation of the multi-factor credit scoring model (Random Forest & Gradient Boosting).
• PyTorch-CreditScoring-ThinFile → Deep learning extension of my thesis (PyTorch, fairness-aware).

• credit-score ↔ Replication Data for Thin File Consumers (Harvard Dataverse).
• Datasets_Review ↔ Alternative Datasets for Thin File Consumers.
• Rural-Credit-Carbon-AI ↔ Unified Dataset for Rural Credit & Carbon Scoring.
• digital-data-literacy-ml ↔ Digital Data Literacy dataset (Dataverse).
• Python-Codes → Exploratory repo for integrating alternative datasets (utility bills, social media).

• AutomaticPlantWateringSystemForArduinoNano → Arduino-based plant watering automation project.
• desktop-tutorial → Default/tutorial repo.

I maintain the DeepaShukla Dataverse – Digital Credit Scoring on Harvard Dataverse:

• Replication Data for: Credit Scoring of Thin File Consumers (2024; modified version 2025 with reproducibility package).
• Replication Data for: Alternative Datasets for Credit Scoring of Thin File Consumers (2024).
• Unified Dataset for Rural Credit & Carbon Scoring (2025).
• Digital Data Literacy Dataset (2025).

All datasets are FAIR-aligned and linked with GitHub repos for reproducibility.

• Machine Learning for Credit & Carbon Scoring
• Fairness-Aware & Interpretable AI
• Alternative Data in FinTech
• Responsible AI & ESG Alignment
• Open Science & Reproducibility