Overview

Before you apply to a job, select your language preference from the options available at the top right of this page.

Responsibilities

• Model Deployment & Productionization
• Deploy ML models into the Global Customer Platform.
• Ensure low-latency inference for real-time decisioning where required.
• Enable scalable batch scoring pipelines.
• Eliminate manual scoring processes through automation.
• Pipeline Automation
• Build automated training and retraining workflows.
• Develop CI/CD pipelines for ML lifecycle management.
• Ensure consistent data refresh cycles aligned with SLA requirements.
• Reduce operational handoffs between Data Science and Engineering teams.
• Model Monitoring & Governance
• Monitor model performance in production environments.
• Detect and mitigate model drift (data drift & concept drift).
• Track prediction accuracy, stability, and bias metrics.
• Maintain versioning and reproducibility standards.
• Feature Engineering & Data Infrastructure
• Design and maintain feature stores.
• Ensure feature consistency between training and inference environments.
• Optimize data pipelines for reliability and scalability.
• Collaborate with data engineering teams on data schema and quality controls.

Qualifications

• 5–10+ years in data engineering, ML engineering, or MLOps roles
• Strong experience deploying ML models into production environments
• Proficiency in Python and ML frameworks (e.g., Scikit-learn, XGBoost, TensorFlow, PyTorch)
• Familiarity with containerization and deployment (Docker, Kubernetes)
• Strong understanding of feature stores and model lifecycle management
• Knowledge of monitoring tools for drift detection and model performance
• Familiarity with CDP integrations and real-time personalization systems
• Understanding of customer segmentation and campaign activation workflows
• Employee Type:
• UPS is committed to providing a workplace free of discrimination, harassment, and retaliation.