Senior Infrastructure Engineer (AI-First)

About MUBI MUBI is a global streaming service, production company and film distributor dedicated to elevating great cinema. To make this possible, we create, curate, acquire and champion visionary films, bringing them to audiences all over the world. We have a team of brilliant, dedicated and passionate people to help bring our mission to life. From London to New York, Istanbul to Paris, and Berlin to Mexico - we work together to realize MUBI’s vision. That’s where you come in! Join our global team and help us make great cinema accessible to everyone, everywhere. About the Role This isn't a traditional DevOps role, and it's not an ML infrastructure role. You're an infrastructure engineer who builds and ships software — internal tools, services, and agentic automation that solve real infrastructure problems. Think automated incident response, intelligent observability, MCP servers that connect infrastructure systems, and support ticket triaging. You apply agentic patterns to the messy operational world, not train models or manage GPUs. We're looking for someone senior — you architect systems, mentor other engineers, and lead by example. You've built real applications and tools in the infra space before, not just configured existing ones. You don't hesitate to poke around unfamiliar systems, take full ownership of problems, and drive them to completion. Our infrastructure team operates on SRE principles: we build reliable, automated systems and eliminate toil. We also build and maintain the platforms that let the rest of engineering deploy, scale, and monitor their services. You'll own problems end-to-end, from bare-metal CDN servers to Kubernetes clusters to the AI-powered workflows that tie everything together. You use AI tools in your own workflow too — not as a novelty, but as a core part of how you build. Small team, high autonomy, direct impact. Hybrid in London (2-3 days/week). Remote considered for strong candidates outside London. Where you'll have impact: Build & Ship — Agentic Infrastructure Design and build agentic workflows for infrastructure operations — incident response, observability, support ticket triaging Build and maintain MCP servers and AI-powered automation that connects infrastructure systems Assess and own the security surface of agentic tools and workflows you build — understand what you're exposing and how to lock it down Build internal tools and services that make the engineering team faster Prototype rapidly using AI-assisted development workflows Collaborate with product engineers — you're a peer, not a support function Write production-quality code, not just glue scripts Platform & Infrastructure Design, run, and evolve our EKS clusters — the primary platform for all MUBI services Build and improve CI/CD pipelines (Jenkins, ArgoCD, Helm) Operate and extend our in-house bare-metal CDN spanning multiple global locations Manage cloud infrastructure on AWS using Terraform and Chef Build and maintain Kubernetes operators and controllers to automate platform workflows Design and operate data pipelines and event-driven architectures using Kafka Maintain and improve observability (ELK, Prometheus, Grafana, Datadog) Define and uphold SLO commitments — own availability and reliability targets Automate everything worth automating Tech stack Cloud: AWS, Kubernetes (currently EKS) IaC: Terraform, Chef CI/CD: Jenkins, ArgoCD, Helm, Kustomize Observability: ELK Stack, Prometheus with Thanos, Grafana, Datadog Streaming/Messaging: Kafka Servers: Nginx (with Lua), bare-metal CDN Databases: MariaDB, PostgreSQL AI/Agents: Claude, MCP, agentic workflow tooling Languages: Language-agnostic — strong fundamentals matter more than specific languages What you'll bring 5+ years in infrastructure, platform engineering, or DevOps Interest or experience in building agentic workflows, AI-powered automation, or working with LLM APIs Strong software engineering skills — you've built and shipped applications and tools in the infrastructure space, not just configured existing systems Strong Kubernetes experience — you understand the internals, not just kubectl apply Deep Linux and networking knowledge — you're comfortable running, debugging, and automating bare-metal servers, not just clicking around a cloud console Security-minded — you think about the attack surface of what you build, especially when integrating AI and agentic tools into infrastructure Deep understanding of distributed systems, high availability, and SLO-driven operations Technical leadership — you architect solutions, mentor engineers, and raise the bar for the team Infrastructure as code experience (Terraform) Nice-to-haves: Experience building Kubernetes operators, controllers, or custom platform abstractions AWS or other cloud provider experience Experience with bare-metal servers or CDN operations Background in backend or full-stack development Ruby, Python, or Go experience Experience with Kafka or event-driven architectures Data pipeline design and operations Experience building with LLM APIs, MCP, or agentic frameworks Familiarity with AI-powered operational tooling (AIOps, intelligent alerting, automated runbooks) Video streaming domain knowledge Don't meet every requirement? Please apply anyway. We're looking for curious, capable people who want to build great things. If you think you'd thrive here, we'd love to talk. We want to make cinema accessible to everyone. We believe people from different backgrounds bring different ideas that foster innovation and engagement, allowing us to attract great people to develop the best experience for our users. MUBI is committed to being an Equal Opportunity Employer. That means it's our responsibility to ensure that all candidates are not discriminated against in our hiring processes and our employment decisions based on their race, color, religion, nationality or ethnic origin, age, gender identity or expression, sex, marital status, physical or mental disability, socioeconomic background, sexual orientation, family or parental status, or any other applicable characteristic.