LMTS Machine Learning Engineer (AI)

TL;DR

LMTS Machine Learning Engineer (AI): Design, build, and productionalize models for customer attrition prediction and mitigation with an accent on scalable data pipelines, feature generation, and ML services integration. Focus on monitoring model performance, drift detection, and collaborating with teams to embed models into agentic experiences and production systems.

Location: New York - New York

Company

Leading cloud-based CRM platform provider powering customer growth and engagement.

What you will do

• Design predictive models for attrition using supervised, unsupervised, deep learning, and generative techniques.
• Build scalable data pipelines for feature generation from product adoption, sales, and engagement data.
• Develop and maintain production-grade ML services with APIs for real-time and batch use cases.
• Monitor and improve model performance via drift detection, retraining, and impact measurement.
• Integrate models into production systems and agentic experiences with engineering and product teams.
• Mentor junior engineers and lead on model architecture, experimentation, and deployment practices.

Requirements

• Proven ability to productionize models from research.
• Strong Python and SQL proficiency for software engineering and data manipulation.
• Experience with ML tools/infrastructure for reusable systems, fast development, and low-latency serving.
• Knowledge of large-scale app architecture: APIs, data pipelines, efficient algorithms.
• Familiarity with ML libraries: scikit-learn, XGBoost, PyTorch, TensorFlow.
• Big data feature engineering with Spark, Trino, Snowflake.
• ML lifecycle tools: MLflow, Airflow, Kubeflow.
• Containerization (Docker) and orchestration (Kubernetes).
• Model evaluation, drift monitoring, explainability best practices.
• Agile, TDD, CI/CD, full software lifecycle ownership.
• Technical communication, mentoring, project management.
• AI Agents integrating with ML models for automated workflows.

Nice to have

• Retention modeling or next best action systems.
• Shared ML frameworks or internal MLOps platforms.
• Feature Stores like Feast.