Support Operations Specialist, AI Agent Management

About Anthropic Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems. About the role: The Support Operations team at Anthropic is dedicated to optimizing and scaling the Product Support function to deliver exceptional user experiences. As a Support Operations Specialist focusing on AI Agent Management, you will sit within our AI Support function and be instrumental in building and improving the capabilities of Fin, our AI support agent. This role sits at the intersection of conversation design and support automation. You'll shape how our AI agent communicates—its tone, structure, handoff logic, and interaction flow—while also building the backend scaffolding that enables it to take real action on behalf of users. The goal isn't just an AI that answers questions; it's an AI that resolves issues end-to-end: processing refunds, managing subscriptions, verifying account details, and routing complex cases to the right humans at the right time. Working as part of our AI Support function, you'll execute on the Fin optimization roadmap—configuring AI behaviors, building integrations that expand what Fin can do, designing conversation flows that create a seamless customer experience, and maintaining the automated workflows that power our support operation. This is a hands-on, technically oriented role where you'll spend your time in Intercom configuring Fin's guidance rules and detection attributes, standing up integrations with tools like Stripe and our status page, and monitoring how these capabilities perform in production. Responsibilities Design and optimize conversation flows, triage logic, and handoff behaviors that create seamless transitions between AI and human support across different customer segments Configure and maintain Fin's Guidance rules and Attributes to improve AI response quality, tone, and inbound categorization Build and maintain integrations between Fin and external systems (Stripe, status page, etc.) that enable the AI agent to take action on behalf of users—processing refunds, fetching subscription details, surfacing incident status, and more Design, implement, and troubleshoot automated workflows for refunds, subscription management, and other transactional support actions, including guardrails, validation logic, and exception handling to ensure safe and reliable execution Own Fin over email setup and maintenance, ensuring consistent AI support quality across channels Contribute to AI support effectiveness monitoring by tracking resolution rates, automation rates, CSAT, and escalation patterns to identify improvement opportunities Provide input on the Fin optimization roadmap based on hands-on experience with configuration, capability gaps, and conversation analysis Partner with the AI Support Systems team on technical dependencies where Fin capabilities intersect with help desk infrastructure You might be a good fit if you: Have 3+ years of experience in support operations, support engineering, chatbot/AI agent configuration, or similar roles in a technology company Have hands-on experience configuring AI agents, chatbots, or automation tools—ideally within Intercom, Zendesk, or similar platforms Think in terms of conversation design: you naturally consider how tone, structure, and flow shape the customer experience, not just whether the right answer gets delivered Are technically comfortable working with integrations, APIs, and workflow builders without needing engineering support for every change Understand that enabling an AI agent to take action requires careful attention to guardrails, edge cases, and safe execution paths—not just connecting systems Demonstrate a data-driven approach, using metrics like automation rate ... (truncated, view full listing at source)