ML Engineer - AI Operations

Location: Morristown, NJ (Onsite) Fulltime

Key responsibilities: Own and operate CI/CD for existing ML services across dev/test/prod; standardize blue/green and canary releases with automated rollbacks. Run model/data drift and performance monitoring with SLAs; define alerts, thresholds, and retraining triggers. Build and maintain production dashboards, alerts, and incident workflows; codify on-call runbooks and escalation paths. Partner with onshore model owners to diagnose metric degradation and land mitigations aligned to governance and controls. Provide day-to-day L2/L3 support for production ML: triage, root-cause analysis, permanent fixes, and post-incident reviews. Own operational documentation: runbooks, standard operating procedures, and recurring health checks. Coordinate hotfixes and safe rollbacks with onshore teams; verify recovery via automated smoke tests. Harden and productionize research notebooks into maintainable, testable services with CI, unit/integration tests, and linting. Operate and evolve model-serving APIs and batch scoring jobs; integrate with enterprise schedulers and data platforms. Ensure models are fully integrated into CI/CD, observability, and monitoring stacks; enforce traceability with experiment and model registries. Validate successful delivery of model outputs to apps, chatbots, reports, and downstream systems with contract tests and data quality checks.

Required Skills: Git/GitLab, Python, SQL, MLflow, Power BI, Snowflake. OLAP/OLTP data modeling and architecture. API frameworks (FastAPI/Flask), and

Nice to have: Modern ELT tools (Fivetran/Airbyte).Streaming/real-time data pipelines (e.g., Kafka, Kinesis, Redpanda).Production ML service operations experience (experience in broader full-stack environments is a plus).