I am a Data Scientist and Machine Learning Engineer with 5 years of experience building predictive models, automating ML workflows, and deploying data solutions. I collaborate with fraud analytics, risk, and supply chain teams to translate complex data into scalable, data-driven decisions. I am proficient in Python, SQL, and modern MLOps tools, with a strong foundation in business analytics and statistical modeling. I enjoy turning messy data into actionable insights and creating transparent, production-ready ML systems that stakeholders can trust.

Ralish Routray

I am a Data Scientist and Machine Learning Engineer with 5 years of experience building predictive models, automating ML workflows, and deploying data solutions. I collaborate with fraud analytics, risk, and supply chain teams to translate complex data into scalable, data-driven decisions. I am proficient in Python, SQL, and modern MLOps tools, with a strong foundation in business analytics and statistical modeling. I enjoy turning messy data into actionable insights and creating transparent, production-ready ML systems that stakeholders can trust.

Available to hire

I am a Data Scientist and Machine Learning Engineer with 5 years of experience building predictive models, automating ML workflows, and deploying data solutions. I collaborate with fraud analytics, risk, and supply chain teams to translate complex data into scalable, data-driven decisions.

I am proficient in Python, SQL, and modern MLOps tools, with a strong foundation in business analytics and statistical modeling. I enjoy turning messy data into actionable insights and creating transparent, production-ready ML systems that stakeholders can trust.

See more

Experience Level

Expert
Expert
Expert
Expert
Expert
Expert

Language

English
Fluent

Work Experience

Machine Learning Engineer (Contract) at JPMorgan Chase & Co.
October 1, 2025 - October 1, 2025
Collaborated with the fraud analytics and risk teams to prototype ML-based anomaly detection models on synthetic and public transaction datasets. Engineered behavioral and temporal features (transaction frequency, time-of-day, merchant category) to enhance fraud pattern recognition. Compared logistic regression, Random Forest, and XGBoost models, improving recall on minority cases while maintaining precision. Used SHAP and LIME to explain model predictions, making outcomes interpretable for non-technical stakeholders. Developed a lightweight Flask API to serve model predictions in real time and integrated visual monitoring dashboards in Plotly. Reported key findings in weekly project standups, helping the analytics team identify potential areas for production integration.
Data Scientist at NEON IT Systems
December 1, 2020 - December 1, 2020
Designed a retail demand forecasting workflow covering data ingestion, feature creation, model training, and deployment. Built forecasting models (XGBoost, ARIMA, LSTM) that reduced average forecast error (MAPE) by 18% compared to previous baselines. Automated ETL processes with Apache Airflow, enabling faster data refresh cycles and improving SLA compliance. Containerized and deployed models using Docker, supporting API-based forecast requests for internal users. Collaborated with supply chain and operations teams to integrate forecasts into planning dashboards, helping reduce stockouts. Created documentation and reusable scripts for model retraining, improving maintainability and team onboarding.
Machine Learning Engineer (Contract) at JPMorgan Chase & Co.
May 20, 2023 - October 20, 2025
Prototyped ML-based anomaly detection models on synthetic and public transaction datasets in collaboration with the fraud analytics and risk teams. Engineered behavioral and temporal features (transaction frequency, time-of-day, merchant category) to enhance fraud pattern recognition. Compared logistic regression, Random Forest, and XGBoost models, improving recall on minority cases while maintaining precision. Used SHAP and LIME to explain model predictions, improving interpretability for non-technical stakeholders. Developed a lightweight Flask API to serve model predictions in real time and integrated Plotly dashboards for visual monitoring. Reported key findings in weekly standups to guide potential production integration.
Data Scientist at NEON IT Systems
June 20, 2018 - December 20, 2020
Designed a retail demand forecasting workflow covering data ingestion, feature creation, model training, and deployment. Built forecasting models (XGBoost, ARIMA, LSTM) that reduced average forecast error (MAPE) by 18% compared to previous baselines. Automated ETL processes with Apache Airflow, enabling faster data refresh cycles and improving SLA compliance. Containerized and deployed models using Docker, supporting API-based forecast requests for internal users. Collaborated with supply chain and operations teams to integrate forecasts into planning dashboards, helping reduce stockouts. Created documentation and reusable scripts for model retraining, improving maintainability and team onboarding.

Education

Master of Science in Business Analytics at University of Texas at Dallas
January 11, 2030 - December 1, 2023
Master of Science in Business Analytics at University of Texas at Dallas
January 11, 2030 - December 1, 2023

Qualifications

Add your qualifications or awards here.

Industry Experience

Financial Services, Retail, Software & Internet, Professional Services, Education