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.
\nFeedback 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.
\nThis 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.
\nYour work will directly impact how fast Anthropic learns from its customers and how reliably that learning shapes what we build next.
\nKey Responsibilities
\nYou'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.
\nYou'll drive adoption through influence, making it so obviously useful that teams pull from it rather than get pushed to it.
\nFeedback Intake & System of Record
\nOwn 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.
\nBuild 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.
\nAI-Enabled Synthesis & Triage
\nBuild 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.
\nDesign the human-in-the-loop model: Claude proposes, PMs and field teams correct, and the system learns from those corrections over time.
\nPartner with Engineering and Research on tooling strategy, evals, and the closed-loop data that makes synthesis quality measurably improve.
\nRouting & Closing the Loop
\nEstablish 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.
\nBuild 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.
\nVoice of the Customer Programs
\nPartner 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.
\nDefine what 'high-signal' means: feedback tied to specific use cases, blocker severity, revenue context, and customer segments so product teams can make confident tradeoffs.
\nContinuous Improvement
\nDefine and track success metrics for feedback loop health , time-to-triage, signal quality, roadmap influence, field satisfaction , and use them to identify bottlenecks.
\nRun regular retros with Product and GTM partners and feed learnings back into process and tooling improvements. Scale what works through documentation and enablement.
\nYou may be a good fit if you:
\nHave 7+ years in product operations, customer insights, voice of the customer programs, or related roles in fast-paced tech companies.
\nHave 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.
\nHave owned a customer feedback program end-to-end , intake, synthesis, routing, and closing the loop , that product teams actually used to make decisions. The customer mix can be enterprise, PLG, design partner, or dev community; what matters is that you designed it and ran it.
\nHave operated at earlier-stage and scaling companies (Series B-D or equivalent) where you built things that didn't exist yet, shipped v1s in weeks not quarters, and iterated in public.
\nHave operated in horizontal, cross-org roles before , you know how to build shared infrastructure that many teams depend on, drive adoption through influence rather than mandate, and earn trust across functions that don't report to you.
\nAre comfortable with ambiguity and can create structure where none exists , you've built the v1 of a system and iterated it into something teams rely on.
\nAre service-oriented and obsessed with making it easy for others to do great work.
\nStrong candidates may also have experience with:
\nBuilding AI-native workflows end-to-end , prompt design, evals, closed-loop improvement , and pushing the boundaries of what automation can own.
\nProduct Management, Customer Success Operations, or Research Operations.
\nFeedback tooling ecosystems (Productboard, Dovetail, or homegrown equivalents) and the tradeoffs between buy vs. build.
\nTreating process as a product with users, metrics, and continuous iteration.
\nTrack record of building and scaling operations programs from zero to one.
\nXML job scraping automation by YubHub
","url":"https://yubhub.co/jobs/job_4249dbdd-13b","directApply":true,"hiringOrganization":{"@type":"Organization","name":"Anthropic","sameAs":"https://www.anthropic.co/","logo":"https://logos.yubhub.co/anthropic.co.png"},"x-apply-url":"https://job-boards.greenhouse.io/anthropic/jobs/5179882008","x-work-arrangement":"hybrid","x-experience-level":"senior","x-job-type":"full-time","x-salary-range":"$260,000-$325,000 USD","x-skills-required":["Product Operations","Customer Insights","Voice of the Customer Programs","AI-Enabled Systems","Process Automation","Collaboration Tools","Data Analysis","Metrics Tracking","Continuous Improvement"],"x-skills-preferred":["LLM Workflows","Prompt Design","Evals","Closed-Loop Improvement","Product Management","Customer Success Operations","Research Operations","Feedback Tooling Ecosystems"],"datePosted":"2026-04-18T15:54:03.763Z","jobLocation":{"@type":"Place","address":{"@type":"PostalAddress","addressLocality":"San Francisco, CA | New York City, NY"}},"employmentType":"FULL_TIME","occupationalCategory":"Engineering","industry":"Technology","skills":"Product Operations, Customer Insights, Voice of the Customer Programs, AI-Enabled Systems, Process Automation, Collaboration Tools, Data Analysis, Metrics Tracking, Continuous Improvement, LLM Workflows, Prompt Design, Evals, Closed-Loop Improvement, Product Management, Customer Success Operations, Research Operations, Feedback Tooling Ecosystems","baseSalary":{"@type":"MonetaryAmount","currency":"USD","value":{"@type":"QuantitativeValue","minValue":260000,"maxValue":325000,"unitText":"YEAR"}}}]}