Senior Tech Lead Platform Engineer

About the role

We build and operate interactive golf games played on screens at driving ranges around the world. Our games are built in Unity with a React-based menu experience, supported by dedicated backends for the frontend, the games, and the monitors themselves — with integrations to booking systems and a data platform built on Kafka and Redshift.

Today, we deploy to a fleet of 40,000+ Linux monitors in 32 countries on a three-week release cycle. We are building a modern fleet management platform — grounded in cloud-native technology, Kubernetes, and platform engineering principles — that will let us provision, update, monitor, and operate every device in our fleet safely, automatically, and continuously. We are using Observe to monitor the health and stability of the system.

As our Senior/Staff Platform Engineer, you will design and build this platform from the ground up. You will own the architecture, make the foundational technology choices, and turn a manual, high-ceremony deployment process into boring, push-button delivery to the edge. This is a rare opportunity to shape a global edge platform at genuine scale — and to build the paved roads that let our game and frontend developers ship without ever filing an ops ticket.

What you will do

• Design and build our fleet management platform for provisioning, updating, and operating 40,000+ Ubuntu-based monitors across 32 countries.

• Own the end-to-end architecture spanning device enrolment, OS image management, application rollout, observability, and remote operations — with a cloud-hosted control plane and lightweight edge agents or clusters.

• Replace our 3-week deployment cycle with continuous, staged delivery using GitOpsand progressive delivery: ring/wave-based rollouts (pilot venues → regions → global), automated canary analysis, and safe automated rollback.

• Build the device and OS lifecycle pipeline covering immutable Ubuntu images, atomic/A-B OTA updates, self-enrolment of new devices, and certificate-based device identity.

• Create self-service golden paths for our development teams so a Unity or React developer can promote a build to a test venue and onward to production without needing to understand the underlying infrastructure.

• Build observability for the whole fleet: using Observe for fleet-wide health dashboards, version and rollout visibility, and alerting that works at 40,000-node scale — integrated with our Kafka-based data platform.

• Solve edge networking for real-world venue conditions: NAT traversal, pull-based management, zero-trust connectivity, and bandwidth-aware artifact distribution (local caching/P2P so a venue with 200 monitors doesn't pull 200 copies of a Unity build).

• Set the technical direction: write RFCs and design docs, drive architectural decisions, and mentor engineers as the platform team grows around you.

What you bring (must-haves)

• 7+ years in infrastructure / DevOps / SRE / platform engineering, with deep, hands-on Kubernetes expertise — including lightweight edge distributions (e.g., K3s, MicroK8s, Talos) and multi-cluster fleet orchestration (e.g., Rancher Fleet, Argo CD ApplicationSets, Flux, Open Cluster Management).

• Experience operating a physically distributed device fleet at scale (hundreds to thousands+ of endpoints), ideally digital signage, retail/POS, kiosks, IoT, automotive, or telco edge — you know what it means when you can't SSH into a device on the other side of the world.

• Strong GitOps and progressive-delivery experience: declarative desired-state management, staged/ring-based rollouts, canary analysis, and automated rollback.

• Solid Linux/Ubuntu endpoint expertise: system administration, image building, and update mechanisms for immutable or atomic systems (e.g., OSTree, Mender, balena, or homegrown equivalents).

• Experience to work integrated with AI agents.

• Cloud and IaC proficiency: designing and operating cloud control planes (AWS preferred), infrastructure as code with Terraform, and CI/CD pipeline design.

• Networking for hostile environments: WireGuard/Tailscale/Nebula-style overlays, pull-based (phone-home) architectures, and operating over unreliable, consumer-grade venue networks.

• Fleet-scale observability: Observe, Prometheus, OpenTelemetry, Grafana — and the judgment to aggregate telemetry from tens of thousands of nodes without drowning in it.

• A genuine platform engineering mindset: you treat internal platforms as products, developers as customers, and you measure success by how rarely anyone needs to talk to you to ship.

Nice to have

• Kafka and/or Rabbit MQ experience (our device telemetry and data platform run on it) and familiarity with Redshift-based analytics.

• Understanding of GPU/graphics workloads on Linux endpoints — drivers, display configuration, and performance monitoring for real-time Unity games.

• Security depth: secure boot, TPM-backed device identity, secrets distribution at fleet scale, and compliance awareness across multiple jurisdictions.

• Experience building or operating an internal developer platform (e.g., Backstage) or comparable self-service tooling.

• Experience introducing platform change across multiple development teams — running migrations, driving adoption, and earning trust.

Who you are

• A pragmatic architect: you know Kubernetes belongs where it adds value, and you can argue for a lighter agent model on device when that's the better answer.

• A strong communicator who writes clear design documents and can explain infrastructure trade-offs to game developers, product managers, and leadership alike.

• Comfortable with ambiguity and ownership — you will define this platform, not inherit it.

• Motivated by removing friction: “reducing deploy anxiety” and “paved roads” are phrases that resonate with you.