Product Operations Manager, Feedback Loops

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 We're hiring a Product Operations Manager — Feedback Loops to own and continuously improve how customer signal flows into product and research decisions at Anthropic. This is a horizontal, org-wide role — you won't be embedded in a single product team, you'll build the shared operating system for voice of the customer that every product team, every surface, and every GTM motion plugs into. Feedback at Anthropic is uniquely high-leverage. We're building on frontier models that evolve constantly, serving customers from individual developers to the largest enterprises, across multiple surfaces (API, claude.ai, Claude Code). Customer signal arrives from everywhere — field conversations, support interactions, early access programs, in-product telemetry — and the opportunity is to make that signal a first-class, structured input to every product and research decision. This role will build the system that makes customer voice as easy to act on as any other data source. You treat feedback loops as a product. You're obsessed with making it effortless for the field to share what they're hearing and for product teams to know what matters most. You build AI-enabled systems that do the first pass so humans can focus on judgment, not triage. You think like a product manager, not a process administrator. Your work will directly impact how fast Anthropic learns from its customers and how reliably that learning shapes what we build next. Key Responsibilities You'll own the operating system for customer feedback across all of Anthropic — one shared platform, not a collection of per-team processes. Working horizontally across every Product team, Research PM, GTM, Customer Success, and Support, you'll establish the intake, synthesis, and routing infrastructure that makes voice of the customer a first-class input to every roadmap. You'll drive adoption through influence, making it so obviously useful that teams pull from it rather than get pushed to it. Feedback Intake & System of Record - Own the single, org-wide pipeline that captures customer feedback from every channel — field teams, support, early access programs, in-product signals — into one structured system of record that serves every product surface. - Build intake workflows that meet teams where they already work (Slack, Gong, CRM) without creating a documentation tax. Obsess over the submitter experience so that sharing feedback is faster than not sharing it. AI-Enabled Synthesis & Triage - Build Claude-powered pipelines that enrich, tag, cluster, and summarize unstructured feedback into trackable issues — doing the first-pass work so humans focus on verification and judgment. - Design the human-in-the-loop model: Claude proposes, PMs and field teams correct, and the system learns from those corrections over time. - Partner with Engineering and Research on tooling strategy, evals, and the closed-loop data that makes synthesis quality measurably improve. Routing & Closing the Loop - Establish clear routing so the right feedback reaches the right product or research owner at the right time — including the path from product signal back into model training priorities. - Build the visibility layer that gives GTM and Support a clear line of sight from customer input to roadmap outcome, so they can close the loop with customers confidently and in real time. Voice of the Customer Programs - Partner deeply with GTM, Customer Success, and Sales to design and run structured voice of the customer programs — customer advisory boards, early access programs, design partner cohorts — that generate high-signal feedback by design. - Define what "high-signal" means: feedback tied to specific use cases, blocker severity, revenue context, and customer segments so product teams can make confident tradeoffs. Continuous Improvement - Define and track success metrics for feedback loop health — time-to-triage, signal quality, roadmap influence, field satisfaction — and use them to identify bottlenecks. - Run regular retros with Product and GTM partners and feed learnings back into process and tooling improvements. Scale what works through documentation and enablement. You may be a good fit if you: - Have 7+ years in product operations, customer insights, voice of the customer programs, or related roles in fast-paced tech companies. - Have personally shipped AI-enabled processes and systems — you've written the prompts, built the evals, and iterated on production LLM workflows yourself. You can talk about model behavior with specificity, not just direct others to build. - Have owned a customer feedback progr

7+ years in product operations, customer insights, voice of the customer programs, or related roles in fast-paced tech companies. Have personally shipped AI-enabled processes and systems — you've written the prompts, built the evals, and iterated on production LLM workflows yourself. You can talk about model behavior with specificity, not just direct others to build. Have owned a customer feedback program.

Own the operating system for customer feedback across all of Anthropic — one shared platform, not a collection of per-team processes. Working horizontally across every Product team, Research PM, GTM, Customer Success, and Support, establish the intake, synthesis, and routing infrastructure that makes voice of the customer a first-class input to every roadmap. Drive adoption through influence, making it so obviously useful that teams pull from it rather than get pushed to it. Build intake workflows that meet teams where they already work (Slack, Gong, CRM) without creating a documentation tax. Obsess over the submitter experience so that sharing feedback is faster than not sharing it. Build Claude-powered pipelines that enrich, tag, cluster, and summarize unstructured feedback into trackable issues. Design the human-in-the-loop model: Claude proposes, PMs and field teams correct, and the system learns from those corrections over time. Partner with Engineering and Research on tooling strategy, evals, and the closed-loop data that makes synthesis quality measurably improve. Establish clear routing so the right feedback reaches the right product or research owner at the right time. Build the visibility layer that gives GTM and Support a clear line of sight from customer input to roadmap outcome. Partner deeply with GTM, Customer Success, and Sales to design and run structured voice of the customer programs — customer advisory boards, early access programs, design partner cohorts. Define what "high-signal" means: feedback tied to specific use cases, blocker severity, revenue context, and customer segments. Define and track success metrics for feedback loop health and run regular retros with Product and GTM partners to scale what works through documentation and enablement.