Skip to content
View MohitPammu's full-sized avatar

Block or report MohitPammu

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
MohitPammu/README.md

Mohit Pammu, MBA

Data Scientist | Business Analytics Expert | MIT Professional Education Graduate

Transforming business challenges into data-driven solutions with 3+ years of cross-functional analytical experience

What I Do

Data scientist with business consulting background and entrepreneurial experience. I combine analytical expertise, operational leadership, and technical skills to build machine learning solutions that drive measurable business impact.

Background: Business analyst → Operations management → Data science, bringing real-world context to technical problem-solving.

Recent Achievements:

  • ~90% accuracy on SVHN digit recognition with 38.6% parameter reduction
  • Built facial emotion detection CNN achieving ~82% test accuracy
  • Delivered targeted marketing strategies through 5M+ ride-sharing data analysis
  • MIT Professional Education Applied Data Science certification (May 2025)

Technical Expertise

Machine Learning & AI
Python TensorFlow Scikit-learn Pandas NumPy

Data Analysis & Visualization
R Power BI SQL Matplotlib Seaborn

Development & Tools
Git GitHub Jupyter VS Code


GitHub Statistics


Featured Projects

SVHN Digit Recognition
CNN Optimization for Edge Computing Applications | ~90% Accuracy
Achieved ~90% accuracy with 38.6% parameter reduction through systematic experimental analysis - essential for deploying ML models in resource-constrained environments. Conducted comprehensive studies on model complexity, learning rate optimization, batch size impact, and AutoML hyperparameter tuning to maximize performance while minimizing computational costs.

Facial Emotion Detection
Customer Experience & Security Analytics | TensorFlow
Developed grayscale-optimized CNN achieving ~82% accuracy for real-time emotion recognition applications in retail analytics, customer service optimization, and security monitoring. Built lightweight, edge-compatible solution enabling cost-effective deployment across multiple business use cases.

Bikeshare User Analytics
Marketing Strategy & Customer Segmentation | R Programming
Analyzed 5M+ ride-sharing records to identify usage patterns between member and casual rider segments. Delivered targeted marketing strategy including user-specific promotions, enhanced member benefits, and membership ranking system to drive customer acquisition and retention.

HR Analytics Dashboard
Workforce Planning & Attrition Prevention | Power BI
Built comprehensive dashboard analyzing 6M+ HR data points to predict employee attrition and optimize workforce planning. Implemented advanced DAX calculations and row-level security, enabling data-driven decisions for talent retention and organizational restructuring.


Education & Certifications

  • Applied Data Science - MIT Professional Education (2025)
  • Data Analytics Professional - Google (2024)
  • Power BI Data Analyst Associate - Microsoft (2024)
  • MBA, Finance Concentration - La Sierra University
  • BA Business Management, Minor: Accounting - La Sierra University

Connect

Portfolio LinkedIn Email


About

Background: Business Analyst (DC consulting, $50M+ healthcare projects) → Operations Manager (restaurant industry) → Data Scientist. I leverage cross-functional experience and stakeholder management skills to deliver technical solutions with clear business impact.

Currently building: End-to-end production ML system demonstrating enterprise-scale capabilities.


Profile views

Pinned Loading

  1. SVHN-Digit-Recognition Public

    Optimized CNN achieving ~90% accuracy with 38.6% parameter reduction for production-ready digit recognition

    HTML 1

  2. Projects Public

    Comprehensive data science portfolio featuring machine learning models, business analytics, and production-ready implementations with quantified business impact

    HTML 1

  3. Facial-Emotion-Detection Public

    Real-time facial emotion detection using optimized CNN architecture achieving ~82% accuracy. Built with TensorFlow/Keras for production deployment with grayscale optimization and data augmentation …

    HTML 1