Product Designer III, AWS Search, Security, Observability, Governance

Amazon OpenSearch Service is how customers search, analyze, and make sense of their data at massive scale — powering application search, log analytics, observability, and increasingly AI-powered and agentic investigation workflows. It's a product where the data is enormous, the queries are complex, and the users range from developers running their first search cluster to SRE and security teams triaging an incident at 2 a.m. We're looking for a Senior Product Designer to own the experience for a core part of OpenSearch and make that complexity feel effortless. Search itself is being reinvented. With Winnow, OpenSearch is moving from keyword-and-query mechanics toward agentic search — where a customer expresses intent and an AI agent plans, retrieves, reasons, and returns answers rather than ten blue links. That shift rewrites the design problem: the interface is no longer a search box and a results list, but a conversation with a system that acts on the customer's behalf, shows its reasoning, and has to earn trust with every answer. You'll help invent what that experience looks like. This isn't a role where design hands off mockups and walks away. Our designers are end-to-end builders — embedded in OpenSearch builder pods, shaping product decisions, prototyping in real code, shipping to production, and measuring what happens. You'll own your problem space from framing through launch, ground your decisions in research that actually feeds the roadmap, and help push OpenSearch into its next chapter — agentic search and observability where the design challenge is as much about trust and verification Key job responsibilities 1. Own the end-to-end design for a core OpenSearch experience — from problem framing and research through interaction design, visual craft, production implementation, and post-launch measurement. 2. Help define the experience for Winnow / agentic search: how a customer expresses intent, how the agent's planning and reasoning are surfaced, how results and their provenance earn trust, and how the customer stays in control. 3. Embed in a builder pod with product and engineering as a true end-to-end builder: shape what gets built and in what order, not just how it looks. 4. Design for genuine complexity — search relevance, vector and semantic search, log analytics, dashboards, observability workflows, alerting, and root-cause investigation — and make dense, high-volume data feel clear and navigable. 5. Help define the experience for agentic observability alongside agentic search, where interfaces surface agent reasoning, findings, and remediation — and where earning the customer's trust is the core design problem. 6. Run research that ships: structured research cycles, trust-calibration studies, and competitive signal capture that feed directly into the roadmap rather than sitting in a deck. 7. Contribute to and extend the OpenSearch design system (OUI) so patterns are consistent, accessible, and scale across the product. 8. Partner daily with PMs and engineers to balance customer needs, deep technical constraints, and business goals. 9. Raise the design bar across the team through critique, mentorship, and reusable patterns. A day in the life Your morning might start with a design review where you present a systems-level concept to product and engineering partners, walking through how a new interaction pattern scales across multiple surfaces. After lunch, you shift into hands-on craft work in Figma, refining a prototype informed by recent research findings. Later, you facilitate an alignment session with cross-functional stakeholders who have competing priorities, helping the group converge on a shared path forward. Throughout it all, you mentor fellow designers, write persuasive strategy documents, and use AI tools to rapidly explore ideas. About the team We're the design team behind Amazon OpenSearch Service, and we work differently: six designers embedded directly in OpenSearch builder pods, owning the experience from problem framing through production. We're designing OpenSearch's shift to agentic search and observability so the work is net-new: inventing the patterns for how customers trust a system that searches, reasons, and acts for them. We operate on three principles: end-to-end builders (we shape, build, ship, and measure), research that ships (structured research and trust-calibration work that feeds the roadmap directly), and a self-learning design system (OUI, evolving toward agentic patterns). We treat design as a competitive weapon, not a line item. We're a senior, tight-knit group that gives designers genuine ownership of their problem space, invests in craft and research, and cares deeply about raising each other's work.