Rajeev Reddy

About

ML engineer/data scientist who deployed a production credit risk + insurance claims triage platform at Hartford Financial, combining XGBoost default prediction with BERT-based document classification. Demonstrated strong MLOps by cutting inference latency to sub-500ms and building drift monitoring plus automated retraining/deployment pipelines (MLflow, CloudWatch, GitHub Actions, SageMaker) with human-in-the-loop review and SHAP-based explainability for underwriting adoption.

Experience

Education

Key Strengths

• Built and deployed production ML system for credit risk + insurance claims triage
• Reduced BERT inference latency from ~3s to <500ms via ONNX + GPU EC2
• Designed drift monitoring and automated retraining (MLflow/CloudWatch/GitHub Actions/SageMaker) on 30–45 day cadence
• Implemented gated deployment with evaluation tests and accuracy/latency SLAs
• Human-in-the-loop process for low-confidence predictions to capture edge cases
• Strong MLOps/orchestration experience (Airflow DAGs, CI/CD, Docker, SageMaker deployments)
• Effective non-technical stakeholder alignment using SHAP explanations and business-defined risk thresholds