Salvia AI

Inspiration

The inspiration for Salvia AI came from a simple but painful reality across many parts of Africa: millions of people live far from hospitals, clinics, or trained healthcare workers. In many rural areas, reaching medical help can take hours, and preventable conditions often become life-threatening simply because people lack early guidance.

We realized that while healthcare infrastructure may be limited, mobile phone access is growing rapidly. Even in low-resource communities, many people already use smartphones or messaging platforms like WhatsApp daily. This inspired us to ask:

What if safe healthcare guidance could reach people instantly, directly from their phones?

That question became the foundation of Salvia AI.

What it does

RuralCare AI is an AI-powered healthcare guidance platform designed specifically for underserved and low-resource communities.

The platform helps users:

Access multilingual healthcare support Describe symptoms and receive safe health guidance Access emergency first-aid information Use an interactive body map for symptom reporting Receive regional outbreak alerts Find nearby clinics and healthcare resources

The goal is not to replace doctors, but to help bridge the healthcare information gap and guide users toward safer decisions.

How we built it

We built the project using a modern full-stack web architecture:

Frontend: React.js, Responsive mobile-first UI, Interactive SVG body map, Multi-language translation system.

Backend: Node.js + Express, MongoDB for conversations and health data, AI integration for symptom guidance, Location-aware healthcare recommendations, and an AI System

The AI system was carefully designed with safety-focused prompting to:

Avoid dangerous medical claims
Encourage professional care when needed
Classify urgency levels

Challenges we ran into

One of the biggest challenges was balancing simplicity with medical safety.

Healthcare information can easily become dangerous if presented incorrectly. We had to carefully design prompts and response structures to avoid misinformation while still making the system genuinely useful. Another major challenge was localization. Most healthcare apps are built primarily for Western audiences, but RuralCare AI needed to work for:

Low-bandwidth environments,
Multilingual communities,
Rural healthcare realities,
And users with varying levels of health literacy.

We also faced technical challenges while building:

the interactive body map, accurate touch/click detection, multilingual translation architecture, and dynamic healthcare categorization systems.

Accomplishments that we're proud of

One of our proudest accomplishments was building a healthcare platform specifically designed for underserved and low-resource communities instead of creating a generic health application.

Another major accomplishment was beginning development of our WhatsApp healthcare assistant, which is live on +1 (415) 523-8886. Since WhatsApp is already widely used across Africa, we recognized that meeting users on platforms they already trust and use daily could dramatically improve accessibility and adoption.

What we learned

We learned:

How critical localization is in healthcare technology,
How to design AI systems responsibly,
How to structure scalable multilingual applications, and how important user trust is in medical interfaces.

We also learned the importance of designing for real-world constraints instead of ideal conditions.

What's next for Salvia AI

WhatsApp voice chat integration
Offline-first support, especially using USSD codes, or even a callable phone number to connect to AI in the rural community.
Voice-based healthcare interaction
Community health worker dashboards
Regional disease surveillance tools
AI-assisted triage systems

Our long-term vision is to create a healthcare guidance platform that is accessible to anyone, regardless of geography or income level.