Senior DevOps Engineer
Core
Deploy, operate, and evolve a microservices-based platform running in Kubernetes clusters across AWS, GCP, and on-prem, including GPU-based ML inference services.
Role type
Senior DevOps/SRE Engineer
Builds
Microservices platform, GPU-based ML inference services, CI/CD pipelines, observability stacks
Domain
Cloud infrastructure, Kubernetes, Machine Learning inference
Deliverable
production ML models
Required skills
Linux system administration, Kubernetes, Terraform, Ansible, Python, GitOps, CI/CD, observability (Grafana, Prometheus, Loki, ELK), Docker, Helm
Preferred skills
ML inference on GPU/CPU, Google Cloud Platform, high-uncertainty environment adaptability, strategic thinking
Technologies
Kubernetes, AWS, GCP, Rancher, RunPod, Scaleway, Nebius, Triton Inference Server, vLLM, GitHub, GitLab, NetBird, PostgreSQL, ClickHouse, Kafka, Superset, Python, Linux, Docker, Terraform, Ansible, Grafana, Prometheus, Loki, ELK
Responsibilities
Deploy, operate, and evolve a microservices-based platform running in Kubernetes clusters across AWS, GCP, and on-prem (Rancher); Operate and support GPU-based ML inference services (Triton Inference Server, vLLM) deployed on RunPod, Scaleway, and Nebius; Build and maintain Docker images for all microservices and ensure a stable service lifecycle; Maintain and scale development and production Kubernetes clusters, actively participate in deployment debugging, incident investigation, and performance troubleshooting; Develop, maintain, and evolve custom Helm charts for each service; Design and operate CI/CD pipelines using GitHub (code and pipelines) and GitLab for on-prem customer deployments; Ensure platform compliance with SOC 2 requirements and actively contribute to improving security and compliance processes; Manage cluster access via NetBird VPN, implementing role-based access control using group policies; Deploy and manage infrastructure using IaC practices with Terraform and Ansible; Develop and continuously improve observability systems: Grafana & Prometheus for metrics, ELK stack for centralized log storage and analysis; Continuously optimize infrastructure in the areas of IaC, IAM, Observability, and CI/CD; Work with a technology stack, including: Python, Kubernetes, Linux, Docker, GitHub CI/CD, PostgreSQL, ClickHouse, Kafka, Superset, Terraform, Ansible