NEIL PARASHAR JHA

Data Analytics and AI professional with over 3 years of experience in delivering impactful, data-driven solutions in business banking operations, workforce analytics, customer experience improvement, and risk management. SQL/ETL workflows (60% faster queries), and developing advanced analytics frameworks for SLA/TAT monitoring, and rework reduction. Skilled in supervised and unsupervised machine learning, NLP for text mining, and generative AI techniques, with a track record of improving operational efficiency—saving approximately $140K AUD annually, reducing rework by 50% year over year, and boosting NPS by 12%. Adept at end-to-end data science workflows in Python, R, SQL, AWS SageMaker, Tableau, and Power BI, and experienced in capacity planning, workforce forecasting, and compliance reporting. Collaborative leader with strong stakeholder management skills, delivering standardized analytics assets that accelerate decision-making across more than 10 teams.

• H2O.ai Certification — AutoML, Machine Learning Models Deployment
• Advanced Analytics Program — Mentem Academy, University of New South Wales (UNSW)
• AWS Data Warrior — CBA

Credit risk default modeling

• Engineered end-to-end pipelines using Python (Pandas, scikit-learn), and SQL to preprocess SME financial and transactional data
• Trained and validated XGBoost and logistic regression models, achieving a 90% ROC AUC and reducing type I errors by 15%
• Identified the top five predictive financial ratios, informing credit policy recommendations adopted by the project sponsor

Customer churn prediction

• Developed feature sets from customer behavior and transaction logs, and addressed class imbalance with SMOTE
• Built and compared logistic regression, random forest, and gradient boosting models, attaining 85% accuracy
• Visualized churn risk segments in Tableau to guide targeted retention campaigns, projecting a 12% lift in customer retention

Time-series forecasting of transaction volume

• Applied ARIMA and Prophet on three years of transaction data stored in AWS S3, achieving 8% MAPE
• Integrated forecasts into a simulation dashboard (Python, Dash) to advise staffing plans that reduced SLA breaches by 15%

• Optimised Business Banking Onboarding via AI-Driven Document Validation & Insight Extraction - Deploy a Gen AI assistant that validates scanned forms and analyses unstructured rework logs, accelerating customer onboarding and improving operational efficiency
• Wilson Mega Hackathon — Home Loan Default Prediction
• Shinkansen Travel Experience Hackathon — Passenger Satisfaction Prediction (92.30% Accuracy)