Senior DevOps Engineer (GCP / ML Platform) IRC306817

Description

GlobalLogic is partnering with a leading global threat intelligence platform to design and deliver an enterprise-scale Human-in-the-Loop (HITL) AI agent platform. The initiative aims to accelerate expert analyst decision-making by combining generative and agentic AI with human expertise.

As a Senior DevOps Engineer (GCP / ML Platform), you will play a key role in building and operating the Google Cloud infrastructure behind this platform. Your responsibilities will include infrastructure as code, deployment automation, model serving, automated retraining pipelines, and secure integration with enterprise data systems. You will also lead production operations and Hypercare, ensuring the platform remains reliable, secure, and scalable after go-live. The initial engagement is planned for 4 months, with the possibility of extension depending on project needs.

Fully remote.

Requirements

• 5+ years of experience in DevOps or platform engineering, including at least 2 years supporting ML workloads in production.
• Strong hands-on experience with Google Cloud Platform (GCP) and Terraform.
• Experience with Vertex AI Pipelines, Endpoints, and Model Registry, or an equivalent MLOps platform.
• Experience building CI/CD pipelines using Cloud Build or GitHub Actions, working with containers, and deploying workloads on Kubernetes.

• Experience implementing enterprise security controls, including IAM, secrets management, and network security perimeters.

  Nice to have:

• Experience with inference optimization techniques, such as ONNX and quantization.
• Familiarity with MLflow, LLM observability tools, and data engineering in BigQuery.

• Relevant Google Cloud certifications, such as Google Cloud Professional-level certifications.

   

Job responsibilities

• Design, provision, and manage cloud infrastructure using Terraform, including Vertex AI, Cloud Run/GKE, Pub/Sub, BigQuery, Cloud SQL, Secret Manager, IAM, and VPC Service Controls.
• Build and maintain CI/CD pipelines for services and ML models, including model registry and experiment tracking capabilities.
• Deploy and optimize ModernBERT model serving to achieve low-latency, cost-efficient inference at scale.
• Automate model retraining based on feedback volume or model drift, with validation and approval gates before promoting models to production.
• Build and operate the data infrastructure supporting analyst feedback and precedent data, using BigQuery and a vector store.
• Implement secure, auditable access to the client’s MCP servers, including authentication, token management, and Cloud Logging.
• Establish observability for system performance, inference latency, cost, and model quality; lead Hypercare, production monitoring, and incident response.