MLOps Engineer (Edge and Accelerated Computing)

GlobalLogic is seeking a specialized MLOps Engineer to develop and manage continuous deployment, optimization, and monitoring infrastructure for Vision-Language Models running on resource-constrained NVIDIA-based automotive platforms. This role focuses on edge computing and accelerated computing environments with expertise in CI/CD automation and model lifecycle management.

• Build and maintain CI/CD workflows that automatically trigger model conversion, quantization, and deployment to target hardware upon code or model updates
• Manage containerized development and execution environments to ensure consistency between high-power training servers and resource-constrained target platforms
• Implement automated monitoring for inference telemetry including latency, throughput, memory pressure, and thermal metrics during automated test runs
• Establish systems for tracking model iterations, ensuring every optimized artifact is tied to training data, quantization parameters, and hardware-specific plugins
• Configure and optimize unified inference serving engines to handle concurrent requests and streaming data on target SoCs

• Proficiency in CI/CD for Machine Learning with expertise in automating complex ML pipelines and Hardware-in-the-Loop testing stages
• Expertise in managing lightweight container runtimes and registry services for ARM64-based automotive platforms
• Experience building auto-profiling scripts that capture low-level hardware metrics using system-level performance diagnostics
• Deep knowledge of experiment tracking and model registry platforms for managing high-dimensional VLM artifacts
• Strong Python and Shell scripting skills for bridging disparate tools in NVIDIA-based software stacks

Position involves working with edge computing and accelerated computing technologies. Remote work arrangements not specified in posting.