ModelOps Engineer

• Lead the integration and deployment of trained AI/ML models into production environments (e.g., cloud, edge devices) using Machine Learning Operations (MLOps) best practices.
• Develop and optimize model training and inference pipelines for real-time execution while efficiently handling large-scale data processing.
• Work with data science teams to structure automated ML model health monitoring, performance tracking, and model refresh capabilities.
• Implement Continuous Integration, Continuous Delivery, and Continuous Training (CI/CD/CT) workflows using commercial and open-source modeling platforms and services.
• Coordinate with Data Science and Data Engineering teams to build scalable feature stores for optimal model training and execution workflows.
• Research, evaluate, and recommend new MLOps engineering tools, applications, and software packages that can be approved and adopted for use in the CBP environment.
• Collaborate with cross-functional teams (e.g., Software Engineering, Data Science) to integrate and test candidate AI/ML models and applications for operational assessment.

• Bachelor’s degree in Computer Science, Data Engineering, Software Engineering, Information Technology, Artificial Intelligence, or a related technical field with 7+ years of experience (or equivalent combination of education and experience, ranging from High School Diploma + 15 years to M.S. + 5 years or PhD + 3 years).
• Experience with MLOps tools, orchestration frameworks (e.g., MLflow, Kubeflow, Apache Airflow, or similar) and/or automated drift detection or monitoring platforms (e.g., Alibi, Grafana, Prometheus).
• Experience with enterprise ML platforms (such as AWS SageMaker, Databricks, DataRobot, or similar cloud-native AI services).
• Experience automating workflow orchestration to manage both batch and real-time streaming data processing for model inference.
• Proficiency in Python, Scala, or Java, along with a strong understanding of high-performance computing, parallel processing, and Graphics Processing Unit (GPU) acceleration.
• Knowledge of productionizing ML models-including optimizing for inference latency and containerization (e.g., Docker, Kubernetes)-or exposure to multi-cloud deployment platforms (such as AWS, Azure, GCP, or comparable environments).

Preferred Qualifications:

• Experience working with distributed data/computing tools and search/indexing platforms (e.g., Apache Spark, Elasticsearch, Solr, Hadoop, Impala, PostgreSQL, or related relational/non-relational data systems).
• Deep understanding of MLOps principles and tools for automated model training, testing, deployment, governance, and continuous monitoring.
• Strong communication skills with a proven ability to collaborate effectively across Data Science, Data Engineering, and DevSecOps teams.
• Experience with data integration and Extract-Transform-Load / Extract-Load-Transform (ETL/ELT) workflows across relational/non-relational databases (such as Oracle, PostgreSQL, MongoDB) and cloud serverless endpoints (e.g., AWS Lambda, GraphQL).
• Experience leveraging deep learning frameworks (such as PyTorch, TensorFlow, or Keras) and computer vision libraries (e.g., OpenCV, SimpleITK, VTK).
• Experience with biometric or image recognition algorithms and associated predictive analytics pipelines.
• Experience managing GPU-based infrastructure and execution performance optimization.

Clearance Requirements:

• Must possess an active Top Secret security clearance OR current DHS CBP Suitability (applicants without active DHS CBP Suitability must hold an active Top Secret or a higher-level clearance in order to be considered).
• Must be able to obtain and maintain full DHS CBP Suitability prior to start.

Physical Requirements:

• The person in this position needs to occasionally move about inside the office to access file cabinets, office machinery, or to communicate with co-workers, management, and customers, which may involve delivering presentations.