Senior AI Infra Engineer: GPUs, CI/CD & Cloud Ops

Mirai, a Scopely company; Riyadh; On-site; SAR 420,000 - 640,000; Full time

Job summary

Mirai, a Scopely company, is building the infra that powers our Generative AI products in Riyadh. This role owns infrastructure as code, CI/CD, observability, and security on AWS, setting the operational standard for a growing team.

You will run GPU-enabled AI workloads, manage container platforms (EKS/ECS/Fargate), API gateways, and secrets, plus build monitoring and incident response into every service.

Strong Terraform/OpenTofu, Ansible, Kubernetes, and GitHub Actions experience are required.

Qualifications

• Eight or more years in DevOps, SRE, or infrastructure engineering with production AI/ML workloads.
• Terraform or OpenTofu with module design and remote state.
• Terraform Cloud (HCP) experience is a plus.
• Ansible for configuration management.
• AWS experience across compute, networking, IAM, and storage.
• GitHub Actions CI/CD with reusable workflows and credentials handling.
• Docker/Kubernetes (EKS) and a registry (ECR).
• API gateways experience (Kong or AWS API Gateway) with auth and routing.
• RDS/Aurora, PostgreSQL, and cache (Redis/ElastiCache) operations.
• HashiCorp Vault or AWS Secrets Manager for secrets.
• Observability tools: Prometheus, Grafana, CloudWatch, OpenTelemetry.
• Linux scripting with Bash and Python.

Responsibilities

• Write and maintain infrastructure as code for reproducible environments.
• Own CI/CD pipelines for building, testing, scanning, and deploying applications and model-serving services.
• Run the container platform (EKS, ECS, Fargate) and GitOps workflows.
• Set up runtime for AI workloads: GPU capacity and model serving (vLLM, Triton, or similar).
• Manage API gateways, networking, load balancing, DNS, and certificates.
• Own secrets, identity, and least-privilege access across environments.
• Operate production databases: clustering, replication, backups, failover, recovery.
• Build monitoring for token usage, GPU utilization, and service level objectives.
• Lead reliability and security: incident response, policy as code, scans, and cost discipline.

Skills

DevOps & SRE

AWS

CI/CD

Observability

Security & IAM

AI workloads

Tools

Terraform/OpenTofu

Ansible

Kubernetes (EKS)

GitHub Actions

Vault / Secrets Manager

Prometheus/Grafana

RDS/Aurora

vLLM/Triton model serving

Job description

Mirai, a Scopely company, is building the infra that powers our Generative AI products in Riyadh. This role owns infrastructure as code, CI/CD, observability, and security on AWS, setting the operational standard for a growing team.

You will run GPU-enabled AI workloads, manage container platforms (EKS/ECS/Fargate), API gateways, and secrets, plus build monitoring and incident response into every service.

Strong Terraform/OpenTofu, Ansible, Kubernetes, and GitHub Actions experience are required.