DevOps engineer in AI infrastructure

Orbitant; United States; Remote; USD 37,000 - 59,000; Full time

Benefits offered by this job

Remote work

Mentoring

MacBook

Job summary

Orbitant ищет DevOps инженеров (mid и senior), работающих удалённо. Вы будете проектировать и разворачивать инфраструктуру как код, управлять Kubernetes кластерами и поддерживать GitOps‑платформу. Важны опыт в CI/CD, наблюдаемость, безопасность и взаимодействие с AI инфраструктурой.

Команда ценит умение наставлять коллег, способность ясно объяснять решения и ориентированность на добавление ценности. Официально удалённая работа с возможностью международного сотрудничества.

Qualifications

• Опыт работы в DevOps, SRE или Platform Engineering на продакшн‑платформах.

Responsibilities

• Проектирование, развёртывание и эксплуатацию инфраструктуры как код (Terraform/OpenTofu).
• Управление Kubernetes кластерами: сеть, авто‑масштабирование, узлы, ingress/egress, стоимость и жизненный цикл.
• Проектирование и поддержка GitOps‑модели развёртывания (ArgoCD/аналоги) и Helm‑библиотеки.
• Разработка и поддержка CI/CD пайплайнов и автономных агентов без статических учётных данных.
• Управление observability (Prometheus, Grafana, Loki) и инцидент‑менеджмент.
• Обеспечение безопасности инфраструктуры и секретов как код.
• Работа с AI инфраструктурой: GPU‑инференс, модели, маршрутизация запросов.
• Оптимизация затрат (FinOps) и управление доступом/идентификацией.
• Наставничество коллег и обмен знаниями внутри команды Orbitant.
• Постоянное улучшение процессов и архитектур платформы.

Skills

Kubernetes

Terraform/OpenTofu

GitOps

CI/CD

Linux fundamentals

Networking

Security

Python/Bash

Observability

FinOps

Tools

ArgoCD

Flux

Helm

GitHub Actions

GitLab CI

Prometheus

Grafana

Loki

Terraform

Job description

Описание:

At Orbitant we are looking for DevOps Engineers (mid and senior, remote). What is Orbitant? Humanity is living through its greatest acceleration. This is not a trend; it is a civilizational inflection point. The companies that will define the next era are not the ones with the most code, but the ones with the best judgment about what to build, how to architect it, and why it matters. Orbitant exists to provide that thrust. We are looking for engineers who don't just write code. They make the decisions that determine what to build and why.

Задачи:

Design, deploy and operate cloud infrastructure managed entirely as code (Terraform/OpenTofu), following best practices and with a long-term view. Own the Kubernetes clusters end to end: networking, autoscaling, node pools, ingress, egress, capacity, cost and lifecycle. Design and maintain the GitOps deployment model (ArgoCD or equivalent) and the Helm chart library used by the product teams. Design and maintain the CI/CD pipelines, including reusable workflows, self-hosted runners and authentication without static credentials. Own the observability platform (Prometheus, Grafana, Loki or equivalents), define actionable alerts, and lead incident response and postmortems. Manage platform identity, access and secrets as code, and harden security in environments with sensitive data. Build and operate AI infrastructure: GPU inference servers, model serving for open-source models, and multi-provider inference routing layers (AI Gateway). Make infrastructure and AI cost visible and governable: attribution by application and team, budgets, quotas and continuous optimisation (FinOps). Take part in designing clean, scalable and maintainable platform architectures alongside other team members. Reduce day-to-day friction for the development teams: templates, internal tooling, living documentation and deployment times. Define platform standards and conventions, and get them adopted without becoming a bottleneck. Lead technically and mentor the DevOps team, driving their professional development. Share knowledge and best practices within the team and across Orbitant. Propose new ideas, technologies or process improvements that add real value. Own the platform you put into production: availability, security, cost and observability.

Требования:

Solid experience in DevOps, SRE or Platform Engineering roles, operating real platforms in production. Hands-on experience with Kubernetes in production: workloads, networking, RBAC, autoscaling, persistent storage and troubleshooting. Strong command of Infrastructure as Code with Terraform or OpenTofu: modularisation, state management and multiple environments. Experience with at least one hyperscaler cloud (Azure, AWS or GCP): networking, identity, secrets management, managed databases and perimeter security. Experience with GitOps and declarative deployment (ArgoCD, Flux or equivalents) and with Helm at the level of writing and maintaining charts, not just installing them. Experience designing CI/CD pipelines (GitHub Actions, GitLab CI or similar) and continuous delivery best practices. Solid fundamentals in Linux, networking (DNS, TLS, routing, VPN) and containers (Docker, multi-stage builds, image security). Experience setting up and running observability (Prometheus, Grafana, Loki or equivalents), defining actionable alerts rather than noise. Applied security culture: secrets management, least privilege, RBAC, certificates and rotation. Scripting and automation skills (Bash, Python or TypeScript/Node) to build internal tooling. Interest and good judgement in AI infrastructure: understanding what serving models involve, governing inference traffic and controlling its cost. Experience leading or mentoring other technical people: support, feedback and professional growth. Ability to work in a team, communicate clearly and explain technical decisions to non-technical audiences. Autonomy and judgement to prioritise, with a platform-owner mindset. An attitude geared towards adding value and seeking continuous improvement. A desire to learn, to teach others and to share knowledge. A level of English that allows you to communicate with international teams and clients.

Будет плюсом:

• Advanced experience with Azure: AKS, Application Gateway/WAF, Azure Firewall, Workload Identity, Entra ID, hub-and-spoke architectures and site-to-site VPN;
• Experience with IaC management platforms such as Scalr, Terraform Cloud/Enterprise, Atlantis or similar;
• Experience with External Secrets Operator, cert-manager, Gateway API, KEDA or Argo Rollouts;
• Experience operating data and messaging on Kubernetes: RabbitMQ, PostgreSQL, PgBouncer, Redis or vector databases;
• Experience with Ansible, Packer and on-premise infrastructure (virtualisation, bare metal, GPU servers);
• Experience serving open-source models in production (vLLM, llama.cpp, SGLang, TGI, TensorRT-LLM) and with the GPU stack (NVIDIA drivers, CUDA, GPUs on Kubernetes): quantisation, KV cache, batching and concurrency/latency tuning;
• Experience with AI gateways or inference proxies (LiteLLM, Azure API Management, Kong AI Gateway, Portkey or similar): multi-provider routing, failover, rate limiting and cost attribution;
• Familiarity with the main providers' APIs (Azure OpenAI, Anthropic, OpenAI) and with LLMOps practices: LLM observability, OpenTelemetry tracing and cost control;
• Experience with hybrid on-premise/cloud inference architectures and with data residency and sensitivity requirements;
• Experience in FinOps and cloud cost optimisation (reservations, rightsizing, spend visibility);
• Experience with regulated environments or with sensitive data and compliance frameworks (GDPR, ENS, ISO 27001);
• Experience defining SLOs, error budgets and SRE practices;
• Experience with AI, or curiosity to learn and work with it day to day (Claude Code, Cursor, MCP servers, agents applied to operations and automation).

Условия:

• 100% remote work (based in Spain)
• Top office in Madrid, drop in whenever you like
• Real flexibility
• Latest-generation MacBook
• Claude Code licence
• Training and mentoring
• Salary between €33,000 and €53,000 gross per year
• Referral bonus
• Real growth opportunities
• Being part of a spectacular team that can't wait for you to join