Platform Engineer: AI Infra & GPU Scale

Ineffable; Greater London; On-site; GBP 90,000 - 130,000; Full time

Job summary

Ineffable is hiring a Platform Engineer for its London office. You will own critical infra, including large-scale GPU compute and developer environments, shaping the platform's future rather than following existing architecture.

The role emphasizes ownership, fast iteration, and building resilient systems to enable ambitious AI research. You'll work on Kubernetes clusters, GPU scheduling, cloud infra, and observability tools while collaborating with a team pushing new paradigms in AI research.

Qualifications

• Experience operating Kubernetes and containerized environments.
• Hands-on with GPU scheduling and large-scale compute.
• Proficiency with Python and Rust for infrastructure tooling.

Responsibilities

• Kubernetes and containers: manage clusters and deploy internal tools.
• GPU scheduling at scale: hands-on with tools like KAI scheduler and Kueue to keep thousands of GPUs busy.
• Making large-scale infrastructure resilient: root causes and cluster-scale chaos prevention.
• Cloud infrastructure: experience on Google Cloud and other major providers.
• Observability that actually helps: log management and monitoring with Grafana or Datadog.
• Developer environments people love: tooling like Tailscale, Workbrew and dev containers.
• Clean, well-crafted code: Python and Rust for tools and infrastructure.

Skills

Kubernetes

Containers

GPU scheduling

Python

Rust

Cloud infra

Observability

Developer tooling

Tools

KAI scheduler

Kueue

Grafana

Datadog

QuickWit

Tailscale

Workbrew

Dev containers

Job description

Ineffable is hiring a Platform Engineer for its London office. You will own critical infra, including large-scale GPU compute and developer environments, shaping the platform's future rather than following existing architecture.

The role emphasizes ownership, fast iteration, and building resilient systems to enable ambitious AI research. You'll work on Kubernetes clusters, GPU scheduling, cloud infra, and observability tools while collaborating with a team pushing new paradigms in AI research.