Product Operations Manager, Human and Agent Experience

SecurityScorecard is a cybersecurity ratings company serving organizations with technology to identify and address cybersecurity risks. The Product Operations Manager will build and run the operating model, cadence, metrics, and standards for a pod-based, AI-first design organization and emerging agent experience discipline. The role will coordinate design operations, AI-enabled workflows, agent evaluations and guardrails, cross-functional planning, and human-agent collaboration.

Responsibilities:

• Design and run the operating model for a pod-based design org: designers embedded in cross-functional product pods, owning continuity per product area, no central ticket queue
• Drive adoption of AI-first design practices: designers prototyping from the live codebase and design system with Claude (or equivalent), shipping working interaction detail instead of static comps
• Own design-to-engineering handoff so interaction details, the details that get lost between Figma and code, are captured, accurate, and buildable on arrival
• Own agent experience (AX) operations: coordinate the evals, guardrails, and agent primitives (permissions, tool access, escalation paths, failure handling) that let agents operate reliably and safely across the product
• Define how humans and agents hand off to each other, and how multiple agents coordinate with each other, so collaboration is designed on purpose rather than emerging by accident
• Define and track the metrics that prove the org is moving faster and improving the overall experience: prototyping velocity, handoff quality, cycle time, agent eval coverage, and AI-tooling adoption on both the human and agent sides
• Run planning, prioritization, and cross-pod cadence, including shared or infrastructure-level design work that spans product lines
• Support customer co-build and live-prototyping sessions, making sure what's learned scales across similar customers instead of staying one-off
• Partner with the UX Director and pod leads to onboard designers into the AI-first workflow and raise the bar on tooling fluency, design judgment, and agentic design skill
• Operate as connective tissue across a flat org, no management layer between designers and the Director, coordinating standards rather than managing headcount

Requirements:

• 5+ years in design/UX operations, program management, or a similar operating role, ideally in B2B SaaS (B2C patterns don't transfer well here)
• Direct experience running AI-enabled design workflows: designers working from a live codebase and design system, producing interactive prototypes, and handing off implementation-ready specs to engineering
• A real point of view on AI-first design-to-engineering workflows, formed by doing it, not just using an AI tool. You can speak to tooling, process, and the tradeoffs you've hit
• Some direct exposure to agent evaluation, guardrails, or agent-facing product design, whether that's building evals, defining agent permission/escalation models, or designing how a product surfaces to an AI agent versus a human
• A track record of replacing centralized, ticket-based intake with embedded, pod-based ownership, or a clear, credible plan for how you'd do it
• Comfort operating without a rigid management hierarchy, driving standards across teams you don't manage
• Strong cross-functional operating skill: you can run planning and prioritization cadences across design, product, engineering, and security without becoming another approval layer
• A bias toward rapid, disposable prototyping over precious, over-invested design work
• We do not provide immigration sponsorship for this position
• Background in cybersecurity, threat intelligence, or another security-adjacent B2B SaaS product
• Experience standing up or scaling an AI-first product development practice at a company going through the same transition (e.g., rolling out agentic coding or design tools org-wide)
• Experience running customer co-build or live-prototyping sessions with enterprise customers
• Experience building or operating multi-agent systems, or defining eval frameworks for agent behavior in production