Staff AI Engineer, Trust & Safety Operations
New York, New YorkFull-TimeStaffAI / Data Science
Responsibilities
- Own the technical roadmap for AI automation across Moderation, Appeals, and Support workflows, driving discovery and prioritization of high-impact AI automation opportunities while providing hands-on technical leadership from concept to production.
- Prototype agentic solutions using the latest platforms and frameworks and integrate them with existing internal and third-party tools and systems.
- Deliver reliable, scalable, and robust automations with the appropriate evaluations, guardrails, human oversight, and clear performance monitoring.
- Drive adoption by producing documentation, running hands-on training and enablement sessions for non-technical operators, and curating prompt libraries and playbooks that empower self-service iteration.
- Collaborate closely with Data Scientists, Data Engineers, Product Managers, Backend Engineers, and the AI Platform Team to ensure a comprehensive and coordinated approach to improving operational efficiency.
- Embed safety, privacy, auditability, and responsible-AI standards into every workflow in partnership with Legal and Security teams.
- Mentor and educate ML Engineers and Platform Engineers on new trends and research in AI/ML that can be applied to Trust & Safety initiatives to promote user safety and improve AI-powered products and workflows.
What We're Looking For
- Agentic & workflow-orchestration expertise: Proven ability to design, build, and operate multi-step LLM agents with modern coordination frameworks.
- Applied AI engineering & prompt craft: Deep Python skills plus hands-on experience integrating foundation models and crafting robust prompts and utilizing vector databases.
- Rapid prototyping & experimentation: Comfortable shipping quick proofs of concept, running A/B or shadow launches, and iterating based on data.
- Backend, data-systems & tool integration: Skilled at wiring external services and internal data into agent workflows through well-designed APIs and schemas.
- Human-in-the-loop system design: Able to blend automation with human oversight through clear escalation paths, review checkpoints, and moderator tooling.
- Operator enablement & training: Talent for translating technical workflows into clear, actionable training for non-technical teams and supporting their day-to-day adoption.
- Working through ambiguity: Proven skill thriving in high-ambiguity, fast-moving environments—prioritizing effectively, adapting plans quickly, and delivering impact amid shifting requirements.
- 7+ years of software or machine-learning engineering experience, with a recent focus on AI-driven automation or agentic systems.
- 2+ years delivering solutions that combine automated decision support with human-in-the-loop review, ideally in Trust & Safety, customer support, or adjacent domains.
- 2+ years designing and tracking operational metrics that demonstrate ROI, accuracy, and user-experience improvements for automated workflows.
- 1+ years of hands-on work prototyping or operating agentic workflows (e.g., MCP, Agentspace, n8n) in real-world or open-source projects.
- A degree in computer science, engineering, or a related field (or equivalent practical experience).
