Key Notes
Production Python services and distributed-system experience are required for this full-time, hybrid London role. You will build the platform that carries AI Driver models through training, simulation, road testing and release. Kubernetes operations and observability are essential; MLOps, approval-based release workflows and React experience are desirable additions.
What You'll Work On
- Build services that automate model progression from feature integration through training, evaluation, approval and promotion across the complete release workflow.
- Integrate model engineering, simulation, measurement and road-testing systems, agreeing interfaces and resolving technical conflicts with the teams that own dependent workflows.
- Improve platform reliability and observability, exposing candidate progress, evaluation results and approvals while using AI-assisted tooling to automate repetitive engineering work.
Why This Role Matters
- AI Driver releases cross several evaluation and operational systems, requiring an identifiable model candidate and consistent approval status as evidence moves between teams and platforms.
- Shared model-release monitoring exposes candidate progress, evaluation results and approvals across dependent systems, allowing teams to identify stalled steps before a model proceeds towards production.
What They Are Looking For
- Technical: Relevant technical capabilities include production Python, distributed systems/APIs and cloud services on Kubernetes.
- Experience: Relevant experience includes service observability, monitoring, alerting and operational reliability.
- Stakeholder: Relevant stakeholder skills include cross-team interface agreement and conflicting-priority resolution.
- Preferred: Preferred attributes include MLOps, model evaluation/release pipelines and React.