As Paragon matures from an experimental identification engine into a production-grade platform, We need dedicated MLOps capabilities to ensure models are deployed reliably, monitored continuously, and retrained efficiently. The Sr. MLOps Engineer will build and own the infrastructure layer that bridges Data Science model development with operational scoring pipelines — ensuring that Paragon's predictions are delivered at scale, with full auditability and drift detection.
Key Responsibilities
- :Design and implement model deployment pipelines from Azure ML to production scoring endpoints, including versioning and rollback capabilitie
- sBuild and maintain model monitoring infrastructure for drift detection, prediction quality tracking, and data quality alertin
- gManage Feature Store operations: pipeline scheduling, freshness SLAs, and serving layer for both batch and near-real-time inferenc
- eIntegrate Paragon scoring outputs into SCOUT operational workflows via the TPL Intel API adapter patter
- nImplement automated retraining triggers based on drift thresholds or scheduled cadence
- sMaintain environment parity across development, staging, and production for all ML workloads on Azur
- eCollaborate with Infrastructure (DevOps) to ensure ML compute environments are right-sized, cost-controlled, and complian
- tBuild observability dashboards for model performance and pipeline health in Azure Monitor or Power B
I
Requirement
- s:4+ years of MLOps or ML infrastructure engineering experien
- ceStrong experience with Azure ML, Azure DevOps, and Python-based pipeline orchestrati
- onFamiliarity with feature stores, model registries, and serving infrastructu
- reExperience in regulated data environments (healthcare, financial services) preferr
- edSolid understanding of CI/CD principles applied to ML workflo
- wsStrong communication skills for async collaboration with U.S.-based Data Science and Engineering tea
ms
