Monogram needs to develop and deploy machine learning models to support operational and clinical use cases, leveraging various data sources including healthcare claims, operational data from care management platforms, and EHRs, to improve patient outcomes and reduce medical costs.
Requirements
- 3+ years of experience using Python for machine learning model development and deployment.
- 2+ years of experience using SQL, Databricks & PySpark to extract and manipulate data for data engineering tasks.
- Experience with MLOps tools and practices (e.g., MLflow, GitHub Actions, Docker, model registries, Azure ML).
- Proficiency in packaging and deploying models in production environments, ideally using Azure cloud services.
- Strong understanding of model monitoring, data drift detection, and model retraining strategies.
- Experience with version control systems (GIT), CI/CD pipelines, and test-driven development.
- Familiarity with cloud computing platforms, preferably Azure, for deploying and managing data science solutions.
Responsibilities
- Own and implement end-to-end ML workflows, including model versioning, testing, containerization, automated deployment pipelines (CI/CD), and post-deployment monitoring for performance and data drift.
- Independently develop and deploy machine learning models to support Monogram’s operational and clinical objectives.
- Conduct data analyses related to model outputs or explore alternative use cases for existing models to enhance reporting and decision-making processes.
- Adhere to and promote best practices in software development, data engineering, and machine learning to ensure high-quality and maintainable code.
- Stay updated with the latest advancements in machine learning and data science, applying new techniques and methodologies to improve model performance and reliability.
- Maintain and update deployed models, ensuring their accuracy and efficiency over time, and troubleshoot any issues that arise during deployment.
- Develop and deploy machine learning models using open-source tools and libraries, operationalized as batch or real-time model inference endpoints.
Other
- Own the development and deployment of machine learning models with minimal supervision.
- Engage with program owners to understand program goals, key performance indicators, and operational details.
- Create and present machine learning solution proposals to program owners.
- 2+ years of experience presenting technical concepts and models to business and executive stakeholders effectively.
- Strong problem-solving abilities and a proactive approach to identifying and addressing business challenges through data-driven solutions.
