Staff DevOps/MLOps Engineer at Sonatus

The Staff DevOps/MLOps Engineer will design and build the foundational, end-to-end DevOps and MLOps platform for Generative AI systems, making critical decisions for large language model-based systems evaluation, monitoring, and deployment. They will implement the full DevOps and MLOps framework, building CI/CD/CT automation to move models from experiment to production with velocity and reliability. Responsibilities include deploying, scaling, and optimizing model serving infrastructure, managing GPU/NPU resources, minimizing inference latency, and building robust monitoring. The role involves creating a cohesive set of best practices for the entire AI lifecycle, defining model versioning, infrastructure as code, and production observability. This includes enforcing reproducibility, approvals, audit trails, PII handling, model cards, and policy/compliance, as well as packaging, versioning, and deploying software modules and AI models with automated retraining.