Description
We're looking for an experienced AI Safety Engineer to drive the deployment and operationalization of automated moderation and guardrail systems that protect our platform and users across a multimodal space. This is a product ownership role where you'll be responsible for end-to-end technical execution of our safety systems, from architecture to deployment and monitoring. You'll bridge the gap between ML research and production-grade systems, ensuring our safety infrastructure is robust, observable, and scalable.
Design and build scalable backend infrastructure for content moderation, abuse detection and agents guardrails, deploying AI/ML models into production systems
Architect robust APIs, data pipelines, and service architectures supporting real-time and batch moderation workflows
Implement comprehensive monitoring, alerting, and observability systems; establish SLIs, SLOs, and performance benchmarks
Partner with ML engineers to translate research models into production-ready systems and integrate them across our product suite
Drive technical decisions and contribute vision to the safety roadmap on how the next generation of platform guardrails should be built for scale and precision.
Requirements:
6+ years of backend software engineering experience building production systems at scale
Strong production backend experience: distributed systems, APIs, data pipelines, and Python expertise (asynchronous Python, backend frameworks)
Infrastructure & DevOps proficiency: cloud platforms (AWS/GCP), containerization (Docker/K8s), CI/CD pipelines
Observability mindset with experience in monitoring tools (Prometheus, Grafana) and building observable systems
Track record of taking products or systems from 0→1 with measurable impact, including deploying or working alongside ML/AI systems in production
Bonus:
Trust & Safety, Content Moderation, or Integrity engineering experience
MLOps experience: deployment, monitoring, and versioning of ML models
Experience with SQL, data analysis tools, real-time streaming systems (Kafka, Redis), or event-driven architectures
Familiarity with React or modern frontend frameworks