MLOps Engineer ID72409

AgileEngine•Brookhaven, GA

5d•Hybrid

About The Position

AgileEngine is seeking an MLOps Engineer to deploy, scale, and monitor computer vision models in production on Azure, processing real-time image streams from thousands of edge cameras. The role owns the transition from AI/ML experimentation to reliable, high-performance production deployment and includes optimizing GPU utilization and infrastructure costs, building ingestion and inference pipelines, and managing model lifecycle and production monitoring. The position is based in Dallas, TX, or requires willingness to relocate there, with office attendance three days per week.

Requirements

Authorized to work for any employer in the United States; the company is unable to sponsor or take over employment visa sponsorship.
At least 3 years of professional experience in MLOps, DevOps, Data Engineering, Machine Learning, or Software Engineering.
Degree in Computer Science, Software Engineering, or a related technical discipline, or equivalent practical experience.
Based in Dallas, TX, or willing to relocate to Dallas; attend the office three days per week.
Experience with production model serving using Triton Inference Server, TorchServe, KServe, or Azure ML endpoints behind REST/gRPC APIs, and end-to-end backend pipeline ownership.
Exposure to experiment tracking, model registries and versioning, drift detection, and production telemetry and monitoring.
Working exposure to computer vision or image data in production.
Strong practical experience with Azure cloud environments, particularly AKS, Azure DevOps, the Azure data stack, and managing or provisioning GPU compute resources.
Deep understanding of containerization, such as Docker and Kubernetes, and designing robust CI/CD pipelines for automated deployments.
Solid conceptual understanding of AI/ML fundamentals to communicate, troubleshoot, and collaborate with applied model developers.
Upper-intermediate English level.

Responsibilities

Optimize GPU compute efficiency and infrastructure costs, maximizing camera throughput per GPU using techniques such as quantization (FP16/INT8), pruning, TensorRT/ONNX conversion, dynamic batching, and GPU autoscaling on AKS.
Own the complete lifecycle transition from AI/ML experimentation to reliable, high-performance production deployment.
Build, maintain, and scale ingestion and inference pipelines for high-volume image streams using Azure Data Factory, cloud storage, and automated CI/CD workflows.
Architect and manage large-scale cloud-based model inference environments, with secondary support for edge GPU deployment for specific use cases.
Implement production monitoring, build visibility dashboards, and set up data and concept drift detection to maintain model accuracy and system reliability.
Manage experiment tracking and model versioning to support reproducibility and traceability of production models.
Collaborate with data scientists and AI researchers to turn experimental models into robust, production-ready solutions.