.png)
Side Effects Story Finder
AI-assisted retrieval and summarization that helps patients find relatable peer experiences fast — without feeling diagnostic.
Overview
Years of community discussions were difficult to navigate. I designed a symptom-and-treatment-based retrieval experience that surfaces relevant peer stories with summarized, digestible insights and explicit trust boundaries.
Problem / Opportunity
What was broken (or missing), and why it mattered.
Threads were valuable but hard to search and slow to skim.
Users had to manually search large archives to find relevant stories.
The product needed to emphasize relatability and reassurance, not recommendations.
My role
Explicit ownership and responsibilities.
- Product discovery
- Search experience strategy
- AI-assisted retrieval planning
- UX definition + IA
- Trust guardrails + language
- Stakeholder communication
Process / Approach
Where the strategy, structure, governance, and tradeoffs live.
- 1) Friction mapping
Identified key breakdowns in community discovery workflows and what “success” looked like for users under stress.
- 2) Structured retrieval interaction
Defined flows for selecting symptoms and treatment context to improve relevance and reduce false matches.
- 3) Summarization workflow
Planned AI summarization to condense long threads into digestible highlights and common themes.
- 4) Trust language + boundaries
Developed language and guardrails to avoid diagnostic framing while promoting emotional support and connection.
Optimized retrieval for contextual matching (symptom + treatment) rather than generic keyword search.
Visual systems diagrams
Placeholders for architecture maps, flows, governance models, and lifecycle diagrams.
A high-level RAG-style pipeline with trust guardrails and UI surfaces.
Technology
Tools and systems involved.
- AI summarization workflows
- Retrieval systems
- Vector search concepts
- Conversational search architecture
- UX workflow design
- Product documentation
- Community content analysis
- Figma
Outcome / Impact
The organizational and product-level effect.
- Created a scalable framework for AI-assisted community retrieval
- Reduced friction navigating large discussion archives
- Improved accessibility of peer experiences within a trusted environment
- Established a foundation for semantic search and storytelling products
Start with an AI Ops audit to identify high-ROI workflows, governance needs, and a pragmatic 30/60/90-day plan.