Dec 2022 – Feb 2024 · 1 yr 2 mos · Ohio, United States · Hybrid

• Designed and deployed credit risk, fraud detection, and predictive analytics models supporting enterprise financial decision-making.
• Built scalable ML pipelines using Python, scikit-learn, TensorFlow, AWS, and Azure for training, validation, and deployment.
• Applied ensemble learning techniques (Random Forest, XGBoost) to improve risk scoring accuracy and fraud detection precision.
• Developed time-series forecasting models to analyze portfolio behavior, financial trends, and market signals.
• Implemented MLOps workflows for automated retraining, model versioning, monitoring, and drift detection using MLflow and Airflow.
• Created Tableau dashboards for near real-time monitoring of risk KPIs and model performance.
• Optimized SQL queries and data extraction processes for large-scale analytics and ML workflows.
• Applied NLP techniques to analyze regulatory and compliance documents for automated insight extraction.
• Ensured data quality, validation, and anomaly detection across production pipelines.
• Mentored junior data scientists and collaborated with engineers to integrate ML models into enterprise applications.

Jan 2021 – Nov 2022 · 1 yr 10 mos · California, United States · Remote

• Developed predictive, statistical, and survival analysis models to analyze population-level health outcomes and program effectiveness.
• Built automated ETL pipelines to ingest and preprocess large public health datasets from multiple sources.
• Applied supervised and unsupervised learning (Random Forest, XGBoost, K-Means, PCA) for healthcare analytics.
• Implemented time-series forecasting models for patient admissions, resource utilization, and demand planning.
• Designed Tableau and Power BI dashboards to visualize healthcare KPIs and program performance for policy stakeholders.
• Applied NLP techniques to extract insights from unstructured healthcare and administrative documents.
• Implemented MLOps workflows using MLflow and Airflow to support model tracking, versioning, and retraining.
• Deployed ML solutions using Docker, Kubernetes, and Azure ML, adhering to government security and compliance standards.
• Collaborated with public health researchers and policy teams to translate analytics into data-driven recommendations.

Jan 2019 – Dec 2020 · 1 yr 11 mos · New York, United States · On-site

• Built predictive models for customer churn, fraud detection, anomaly detection, and usage pattern analysis on large telecom datasets.
• Developed machine learning models using Python, scikit-learn, and ensemble methods to improve churn prediction and targeting.
• Designed Tableau dashboards tracking telecom KPIs such as churn, ARPU, network performance, and usage trends.
• Built automated ETL pipelines using Python and SQL to ingest and validate customer and network data.
• Applied time-series forecasting for network load, call volumes, and revenue projections.
• Conducted A/B testing and statistical analysis to measure impact of pricing and service changes.
• Supported production deployment and monitoring of ML models using Docker, MLflow, and Airflow.
• Collaborated with network operations and IT teams to integrate analytics into enterprise systems.

Aug 2014 – Feb 2018 · 3 yrs 6 mos · India · On-site

• Performed data extraction, cleaning, and preprocessing using SQL and Python to support analytics initiatives.
• Built automated ETL workflows and dashboards using Tableau and Power BI for operational reporting.
• Conducted EDA, trend analysis, and statistical testing to identify business insights.
• Assisted in building predictive models for churn, customer segmentation, and usage forecasting.
• Developed Python automation scripts to streamline recurring reports and data updates.
• Built clustering and segmentation models to analyze customer behavior patterns.
• Collaborated with cross-functional teams to support data quality, analytics, and reporting needs.