AI Knowledge Architect

AI Knowledge Architect Department: Community Operations Team: Community Operations Location: Chicago, United States Type: Full-Time **About the Role** The AI Knowledge Architect (customer support) role sits within the Global Digital Experience (GDX) team and is responsible for building and maintaining the knowledge portion of the brain of our customer-facing AI agents. You will develop the knowledge needed by LLMs to craft empathetic responses, design conversational logic, and provide business context that allow our AI agents to respond against complex customer problems autonomously. You will be the connector between expressed customer requests, support policies and customer data and convert these into intuitive, empathetic and brand-aligned AI responses. Your role requires a combination of skills - precision of a Technical Writer, empathy of a UX Designer, and the logic of a Prompt Engineer. Your role starts with writing articles but extends to structuring knowledge so an AI agent can reason, retrieve, and act upon it accurately. This role sits at the intersection of Community Operations (CommOps) and Customer Obsession (CO) Product. You will work closely with CommOps frontline teams to deeply understand their customer requirements and collaborate with engineering and product teams to build knowledge that optimizes segmented customer experience. **What the Candidate Will Do** - **AI Knowledge authoring:** Author and structure prompt-aware content for AI agents to ensure the AI provides accurate, empathetic responses without hallucinations. Decide where the AI does not need any additional content, where it needs some context, where it can pull data from upstream systems for context and where it needs a lot of details, and tailor knowledge accordingly. Incorporate tribal knowledge from various parts of Uber into the knowledge base to ensure the AI agents are able to respond to customer queries effectively. - **AI Information architecture:** Ensure the content provided to the AI agents are relevant, accurate, updated, non-conflicting and succinct. Identify gaps in content leading to AI agents