DevOps Engineer

Job Description

We re seeking a DevOps Engineer with hands-on experience managing GPU infrastructure, Kubernetes, and hybrid cloud environments (bare-metal, AWS, GCP). You ll work closely with AI researchers and full-stack developers to build and scale the infrastructure that powers our image-processing microservices. Responsibilities Manage and optimize Kubernetes clusters across bare-metal servers, AWS (EKS), and GCP (GKE) Deploy and maintain GPU-enabled workloads for AI inference and training (NVIDIA drivers, nvidia-docker, MIG configs) Create and maintain CI/CD pipelines (GitHub Actions, ArgoCD, etc.) to automate deployments and model rollouts Implement scalable, fault-tolerant infrastructure for AI microservices, using Celery, Redis, and FastAPI Monitor system performance, resource utilization (CPU/GPU), and model latency Set up and manage persistent storage (MinIO, S3), secrets, and config maps securely Develop monitoring and alerting systems for both infrastructure and AI pipelines Collaborate with AI engineers to support experimentation, benchmarking, and model updates Required Skills Solid experience with Kubernetes, particularly in GPU scheduling and resource management Experience deploying and tuning AI/ML workloads on GPUs (NVIDIA Docker, CUDA stack, drivers) Comfortable managing hybrid cloud infrastructure: bare-metal servers, AWS, and GCP Deep knowledge of Docker, Helm, Strong scripting skills (Bash, Python) for automation and tooling Experience with Redis, Celery, and handling message queues or background job systems Tech Stack Infra: Docker, Kubernetes, Helm, Terraform, GitHub Actions Cloud: AWS (EKS, EC2, S3), GCP (GKE, Compute), Bare-Metal Servers AI Ops: NVIDIA Docker, CUDA, Celery, Redis, FastAPI Storage: MinIO, AWS S3, Persistent Volumes