Sage+: Generative AI for Knowledge Discovery
Client: AWS-Amazon Software Builder Experience | Project length: 2023-2025 | Tools: Figma, Adobe Illustrator
AI | 0 → 1 | Design Leadership | Mixed-Method Research | End-to-end UX
In 2023, knowledge discovery at Amazon was a known pain point for software engineers. Finding a reliable answer meant knowing where to look, who to ask, and whether the information was still current. The Knowledge Tech team set out to build their own LLM-powered service to fix that. I was brought on as the sole UX designer to build the experience around it, from the ground up, across five surfaces over two years.
Design & Impact
What I Designed
Internal Search: Implicit+explicit experience
Integrating Sage+ into Search introduced new complexity. The experience offered an automatic AI response on any keyword search alongside the option for explicit follow-up. I designed a non-disruptive loading pattern and a user-controlled toggle to opt-in to automatic responses after engagement data revealed a clear split in user preference.
My Approach
Sage (Q&A): Implicit experience
Sage, the internal question-and-answer platform, was the first product to offer Sage+. AI responses surfaced within two minutes of posting a question. I established the foundational research that supported all future efforts and a visual identity that carried through multiple surfaces.
Sage+ on Sage
Sage+ on Internal Serach
Slack: Explicit experience
The third stage of Sage+ brought an explicit chat experience in a conversational context to users, launching first on Slack. I defined emoji language patterns and scoping controls specific to the way users interact with Slack for in-the-moment problems.
DevCon 2025
DevCon: research at scale
When the team was invited to Amazon's internal developer conference with 7,000+ attendees, I managed and produced the booth content and materials, and turned it into a light research opportunity. Informal survey posters sparked conversations and generated qualitative signal the team couldn't have collected at this scale.
Slack
Wiki: Implicit, explicit and chat experiences
Wiki offered us the opportunity to offer all three experiences: implicit, explicit, and chat. At the page level, users can now scope responses to the current page, the page and its children, or the full Sage+ knowledge base, a meaningful difference for the complexity of questions engineers bring to documentation.
Wiki - Wiki combined all three experiences: Implicit, Explicit and Chat
Selected Moments
Building a brand to drive adoption
As Sage+ expanded to new surfaces, I identified that the product needed a visual identity to stand out from other AI products competing for the same internal users. I ran a competitive analysis of existing generative AI tools and created a brand that gave Sage+ a distinct and consistent presence across Sage, Slack, and beyond.
Sage+ visuals
See it live
The following experiences are accessible to Amazon employees only.
Sage+ on Slack: Type “@Sage” to install the app and start a conversation
“Somewhat helpful” feedback addition
Design that moved an annual goal
Positive feedback had plateaued. Qualitative research revealed that binary good/bad feedback was too limiting for AI responses that were partially correct. Staring with Internal Search, I designed a "somewhat helpful" option, which produced a 10% lift in positive feedback and helped the team meet their annual target.
Wiki: Implicit, explicit and chat experiences