Senior Machine Learning Operations Engineer

Key Skills

Apache Airflow, Azure, Continuous Integration & Continuous Delivery - CI/CD, Data pipelines, Databricks, DevOps, Git, Github, Grafana, Prometheus, Snowflake

Job Description

Job Scope/Complexity

• The Senior Machine Learning Operations Engineer will be part of the Johns Hopkins Medicine Technology Innovation Center and will work full-time with the Precision Medicine team (as well as clients, data stewards, project/program managers, and other IT teams). This position will be a critical member of the Precision Medicine initiative’s data science team, helping to translate new science into improved operations and clinical practice across a range of health conditions.

Specific Duties & Responsibilities

• Develops, deploys, and maintains machine learning models and systems.
• Collaborates with data scientists, software engineers, and other stakeholders to develop, test, and improve machine learning models.
• Designs and implements infrastructure and data preparation processes to support machine learning development and deployment.
• Monitors and optimizes machine learning performance, including model accuracy, efficiency, and scalability.
• Develops and maintains machine learning pipelines, including data ingestion, preprocessing, feature engineering, model training, and deployment.
• Builds monitoring and altering systems for pipeline metrics and model results.
• Identifies and resolves issues related to machine learning model performance and data quality.
• Develops and maintains documentation for machine learning systems and processes.
• Provides customer support to data scientists who use the processes and tools described above.

Scale/Size of Area, Project and/or System Supported

• Precision Medicine Analytics Platform
• Developing pipelines and deploying algorithms within the hospital administration and

• operations segment of the Platform, including regulatory and hospital finance.

• Precision Medicine Delivery Platform
• Developing pipelines and deploying algorithms into clinical workflows.

Special Knowledge, Skills, & Abilities

• Experience with machine learning frameworks such as TensorFlow, PyTorch, or Scikit-learn.
• Microsoft Azure and/or Amazon Web Services cloud management and administration including environment provisioning, configuration, performance monitoring and security.
• Experience with containers and container orchestration using Kubernetes (Docker, Helm, kubectl, linkerd).
• Experience with data engineering tools and languages such as Apache Spark, Python, Java, and Scala.
• Proven track record of delivering AI/ML deployments and cloud-native applications using Infrastructure-as-Code (IaC) and GitOps technologies (e.g., Terraform, ARM templates, Bicep).
• Excellent problem-solving and analytical skills.
• Strong written and verbal communication skills.
• Ability to work independently and in a team environment.

Work Location:

• Remote or Hybrid, East Baltimore

About the Team

The Precision Medicine Initiative is a strategic priority for the Johns Hopkins University and School of Medicine. We are combining data from around the institution, to make it possible for researchers from medicine, operations, and health economics to discover potential new healthcare interventions. We also help them build a path for implementation of these interventions. We are doing this at scale and across a broad range of disciplines.

The TIC is uniquely positioned within Johns Hopkins to help design, build, and deploy novel enterprise, departmental, and clinical applications that strive to improve workflow, outcomes, and patient care. Our applications focus on turning data and algorithms into useful visualization tools, improving workflows such as patient hand-off, rounding, and cross-team collaboration, and helping patients with medication adherence, data gathering, and overall clinical education. Our team frequently works with data from multiple sources including our Enterprise EMR (Epic), researcher curated data, and other IT systems. The applications we design, build, and support are critical to our healthcare operations and require high-availability, 24/7 monitoring, and on-call support.

Minimum Qualifications

• Bachelor's Degree.
• Six years of related experience.
• Additional education may substitute for required experience and additional related experience may substitute for required education, to the extent permitted by the JHU equivalency formula.

Preferred Qualifications

• Bachelor's or Master's Degree in computer science, Data Science, or a related field.
• At least 3 years of experience in machine learning development, model testing, and deployment.
• Experience setting up operations and pipelines for new and existing projects and/or familiarity with version control (i.e., Git, GitHub), CI/CD tooling (e.g., Azure DevOps Pipelines artifact repositories (e.g., Azure Container Registry, JFrog Artifactory), monitoring tools (e.g., Grafana, Prometheus), model monitoring tools (e.g. Weights & Biases, mlflow), and associated process automation (e.g. Apache Airflow).
• Familiarity with complex Data Lakehouse and data management tools, standards, and techniques that support various modes of data ingestion and curation such as: Azure Data Factory, Azure Data Lake Storage Gen2 (ADLS), Azure Databricks, Delta Lake, Azure Event Hubs, and Snowflake.
• Experience in healthcare IT.
• Familiarity with healthcare data standards including Fast Healthcare Interoperability Resources (FHIR), Observational Medical Outcomes Partnership (OMOP) and Observational Health Data Sciences and Informatics (ODHSI) Common Data Model, and/or HL7 Clinical Document Architecture (CDA).