Niteesh P
Data Scientist @Berkshire Hathaway
Signup · Get unlimited contacts
WORK HISTORY
Data Scientist @Berkshire Hathaway
Engineered credit risk models (PD, LGD, EAD) using gradient boosting on 50M+ records, increasing prediction accuracy by 22% and optimizing AUC by 15% to support lending decisions for a $2.5B portfolio.• Quantified key risk drivers using statistical analysis and feature importance, reducing default risk exposure by 18% and strengthening regulatory compliance strategies.• Implemented fraud detection systems using ensemble and neural network techniques on 10M+ transactions, increasing detection rate by 35% and lowering false positives by 18%.• Generated customer segmentation and lifetime value models across 5M+ users, boosting marketing ROI by 27% and increasing retention rates by 15%.• Accelerated model training workflows on AWS, reducing processing time by 40% and improving experimentation cycles for analytics teams.• Delivered insights through dashboards tracking 15+ KPIs, reducing decision turnaround time by 30% for business stakeholders.
EDUCATION
University of Central Missouri
Master of Science - MS, Computer Science
Lakireddy Bali Reddy College of Engineering(Autonomous)
Bachelor of Science - BS, ECE
ABOUT NITEESH P
Data Scientist with 5+ years of experience building scalable machine learning solutions across financial services and analytics.Currently at Berkshire Hathaway, where I design and deploy production-grade ML models for credit risk and fraud detection.Key Achievements:• Improved credit risk model accuracy by 22% using XGBoost• Built ML pipelines processing 2TB+ data daily using PySpark & AWS• Increased fraud detection rates while reducing false positivesTechnical Expertise:Python | SQL | PySpark | XGBoost | TensorFlow | AWS | MLOps | NLP | A/B TestingI specialize in building end-to-end ML systems that drive measurable business impact across risk, fraud, and customer analytics.Currently open to Data Scientist / Machine Learning Engineer roles.Let’s connect!
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.