Gaurav Kaushik
Data Scientist Senior Consultant @Fractal
Signup · Get unlimited contacts
WORK HISTORY
Data Scientist Senior Consultant @Fractal
Gurugram, IN
Developed and deployed machine learning models for product interest prediction using classification algorithms (XGBoost, Random Forest, Decision Trees) in Python and Scikit-learn on AWS SageMaker, improving model accuracy and enabling data-driven product recommendations.• Built comprehensive model monitoring and MLOps framework to detect data drift (PSI/CSI metrics), ensuring model reliability and performance in production environments.• Designed reusable feature store architecture to accelerate data science workflows, standardize feature engineering, and enable rapid experimentation across ML teams.• Partnered with cross-functional stakeholders to define analytics roadmap, experiment design, KPI frameworks, and business intelligence requirements.
EDUCATION
Army Institute of Technology (AIT), Pune
Bachelor of Engineering - BE, Computer Engineering
Great Lakes Institute of Management
Post Graduate Program, Machine Learning
SKILLS
ABOUT GAURAV KAUSHIK
I am Data Scientist / Senior Consultant specializing in the \"Last Mile\" of Analytics & ML—bridging the gap between raw data foundations and production-grade analytics systems. My career has evolved from building foundational Business Intelligence at scale to developing high-impact ML models for some of the world\'s largest organizations.The Evolution:• Phase 1 (BI & Foundation): Architected intelligence systems for Fortune 100s, reducing reporting cycles by 75%•Phase 2 (DS & Modeling): Developed risk and premium model (XGBoost, GLMs) for the US P&C Insurance market, driving measurable lifts in conversion and premium accuracy.Phase 3 (MLE & Monitoring): Building Product Interest Modeling for a global SaaS leader, partnering with MLE teams to deploy and monitor models (PSI/CSI) at massive scale.Technical Advocacy:I am a contributor to the DS community on Medium, where I write about the technical internals of XGBoost, Data Science and Model Monitoring. I believe that a model is only as good as its monitoring framework.Core Stack: Python, Snowflake, PySpark, SQL, XGBoost, Tableau.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.