Remote AI Safety Engineer - Production Guardrails

ElevenLabs; Greater London; Hybrid; GBP 110,000 - 160,000; Full time

Benefits offered by this job

Innovative culture

Growth paths

Learning & development

Social travel

Annual company offsite

Co-working

Job summary

ElevenLabs is seeking an experienced AI Safety Engineer to deploy automated moderation and guardrails across a multimodal platform. You’ll work with ML and full‑stack engineers to build production‑grade safety infrastructure from the ground up.

This product‑ownership role requires end‑to‑end execution from architecture to deployment and monitoring, bridging ML research and scalable safety systems with strong observability.

Qualifications

• 6+ years of backend software engineering experience.
• Experience tackling free-tier abuse, fraud detection, or bot farming at scale.
• Strong production backend experience: distributed systems, APIs, data pipelines, and Python (asynchronous Python).
• Infrastructure & DevOps: cloud platforms (AWS/GCP), Docker/K8s, CI/CD pipelines.
• Observability mindset with experience in monitoring tools (Prometheus, Grafana).
• Experience deploying or working with ML/AI systems in production.

Responsibilities

• Design and build scalable backend infrastructure for abuse detection and production-ready safety systems.
• Architect robust APIs, data pipelines, and service architectures for real-time and batch moderation workflows.
• Implement monitoring, alerting, and observability with SLIs/SLOs and benchmarks.
• Partner with ML engineers to translate research models into production-ready safety systems.
• Drive decisions and contribute to the safety roadmap for scalable, precise guardrails.

Skills

Backend

Python

AWS

Docker

Kubernetes

Observability

ML/AI systems

Tools

Prometheus

Grafana

Kafka

Redis

CI/CD

Job description

ElevenLabs is seeking an experienced AI Safety Engineer to deploy automated moderation and guardrails across a multimodal platform. You’ll work with ML and full‑stack engineers to build production‑grade safety infrastructure from the ground up.

This product‑ownership role requires end‑to‑end execution from architecture to deployment and monitoring, bridging ML research and scalable safety systems with strong observability.