ST Wayfinder: Smile Train
The Arizona State University Artificial Intelligence Cloud Innovation Center, powered by Amazon Web Services (AWS), collaborated with Smile Train, the world’s largest cleft-focused organization, to design Smile Train Wayfinder - a location-based chatbot that helps patients and families quickly find cleft care near them through a simple conversational experience. Smile Train Wayfinder supports users worldwide by returning nearby treatment center details (name, address, and local contact information) based on city, postal code, or general region, especially valuable for underserved or remote communities.
Problem
Smile Train supports cleft care through a sustainable local-partner model across 75+ countries, but many cleft-affected individuals and families still struggle to locate appropriate care quickly - often due to awareness gaps and difficulty navigating resources.
Key challenges include:
- Limited visibility into nearby Smile Train partner treatment centers in many regions
- Time-consuming searching for reliable, location-specific care information
- Increased barriers for users in underserved or remote areas who need fast, clear guidance
Student Spotlight
Role: Full stack developer | Connect with Jenitha on LinkedIn
Experience: As the full-stack developer on this project, I built the end-to-end architecture for Smile Train Wayfinder, an AI-powered multilingual chatbot that helps patients and caregivers locate the nearest cleft treatment centers anywhere in the world. Users simply enter their city, postal code, or region and receive a tailored list of nearby cleft care providers — breaking down geographic and language barriers for families seeking critical care.On the backend, I integrated AWS Bedrock with Claude 4.0 as the conversational AI engine, leveraging its multilingual capabilities to serve a global user base. I designed and implemented the full API layer, built the location-based parsing and query logic, and connected everything to the supporting AWS infrastructure for data storage and retrieval. On the frontend side, I brought UI/UX designer Lahari's high-fidelity Smile Train-branded designs to life as a fully functional, responsive application.
This project gave me meaningful hands-on experience architecting a real-world healthcare tool on AWS, and I'm proud to have contributed to something that could genuinely help patients access life-changing cleft care resources.
Role: UI/UX Designer | Connect with Shakthi on LinkedIn
Experience: As the UI/UX designer in this project, I had the wonderful opportunity to design a multilingual chatbot, Smile Train Wayfinder, that helps individuals find cleft care all over the world. By simply entering their city, postal code, or region, users will receive a tailored list of nearby cleft treatment centers. I used the branding guidelines provided by Smile Train as well as the Smile Train website and project scope documentation to create low-fidelity & high fidelity designs for the chatbot. I’m very proud of the clean, easy-to-use interface I created, and how it will come across to users as a chatbot that is here to help them find resources just with their location.
Approach
The CIC team built Smile Train Wayfinder as a serverless, multilingual service-locator chatbot on AWS with three main flows: Visitor Chat, Admin Dashboard, and Knowledge Ingestion.
- AI/RAG: Amazon Bedrock powers the conversational experience and retrieval over Smile Train resources and partner datasets (including Smile Train India and partner JSON sources) to return accurate, location-based treatment center details.
- Chat streaming: AWS Lambda streams chatbot responses through Amazon API Gateway to deliver a fast, conversational experience for users searching by city, postal code, or region (with a zip/postal-code prompt when location permissions aren’t granted).
- Content ingestion: Amazon S3 stores knowledge assets, including Smile Train web resources and partner datasets (e.g., Smile Train India and Smile Train Express partner JSON files) used to power the service locator and related FAQs.
- Frontend: A web-based UI provides an intuitive search flow aligned to Smile Train branding and supports multilingual access across English, Hindi, Bengali, Telugu, Marathi, Tamil, Urdu, Gujarati, Kannada, Odia, and Malayalam.
- Admin & analytics: Cognito-secured admin access and a lightweight dashboard enable visibility into demand—especially common request locations—along with intelligent conversation summaries that can be shared with stakeholders.
- Infrastructure: AWS CDK deploys the fully serverless stack (Lambda, API Gateway, S3, and supporting services) for scalability and cost-efficient global access.
Industry Impact and Problem Solving
Smile Train Wayfinder reduces friction in accessing cleft care by turning a complex “where do I go?” search into a clear conversational path.
The solution helps by:
- Connecting patients and caregivers to nearby cleft care options faster using location-based recommendations
- Improving access for underserved communities by simplifying discovery of free, high-quality care
- Enabling Smile Train staff to better understand demand patterns through aggregated request insights
“Working with ASU was smooth, collaborative, and genuinely productive. The team asked the right questions, understood our goals, and created a chatbot experience that was useful, thoughtful, and mission aligned. We were especially impressed by the focus on helping people find care in ways that respected different comfort levels around location sharing, along with added features like administrative controls and a cost-effective, scalable technical approach.”
Peter Allala, IT Systems Administrator
Potential for Wider Application
While designed as a cleft care locator for Smile Train’s global network, the same approach can extend to other location-based support needs, including:
- Service navigation for additional healthcare programs and outreach initiatives
- Multilingual resource discovery for remote and underserved populations
- Expanded informational queries (e.g., region-level procedure insights and recent activity) as data integrations mature
Supporting Artifacts
| Github Link: | Click Here |
Next Steps
Our next steps are to evaluate how the ideas and capabilities developed through Smile Train Wayfinder can inform a future evolution of tools already in production, with a focus on broader usability, scalability, and more thoughtful use of AWS services. This project reinforced for us that AI can be most impactful when it is grounded in trusted data and designed around real human needs. It also expanded our thinking about how conversational AI, multilingual access, and flexible location-based experiences can help more patients and families discover care more easily and confidently.
About the ASU CIC
The ASU Artificial Intelligence Cloud Innovation Center (AI CIC), powered by AWS is a no-cost design thinking and rapid prototyping shop dedicated to bridging the digital divide and driving innovation in the nonprofit, healthcare, education, and government sectors.
Our expert team harnesses Amazon’s pioneering approach to dive deep into high-priority pain points, meticulously define challenges, and craft strategic solutions. We collaborate with AWS solutions architects and talented student workers to develop tailored prototypes showcasing how advanced technology can tackle a wide range of operational and mission-related challenges.
Discover how we use technology to drive innovation. Visit our website at ASU AI CIC or contact us directly at [email protected].