RAKTARPAN

More than a gift, it's a lifeline

Comment 2

Inspiration

The daily struggle of Thalassemia and their families to find a regular, reliable supply of blood is immense. Our inspiration came from wanting to solve this dual challenge: not only the logistical problem of connecting donors but also the accessibility problem, ensuring that anyone, with or without a smartphone, can be part of this life-saving network in India.

What it does

RAKTARPAN is a hybrid, AI-powered platform designed to create a reliable community of blood donors for Thalassemia patients.

  • Connects patients in need with nearby, eligible donors in real-time.
  • Provides universal access through two channels: a simple IVR phone helpline for non-tech users and a user-friendly mobile app.
  • Uses AI to predict blood demand in specific areas and to identify donors who are most likely to be available, preventing "donor fatigue."
  • Encourages recurring donations through a meaningful recognition program that honors donors as "life-savers."

How we built it

We designed the architecture for RAKTARPAN using a modern, scalable tech stack.

  • Platform: Python (Django) backend for robust logic and a React Native mobile app for cross-platform reach.
  • AI Engine: Designed with Scikit-learn & XGBoost for predictive donor availability and demand forecasting.
  • Accessibility: Twilio for the multi-lingual IVR helpline and real-time SMS alerts.
  • Database: PostgreSQL to securely manage sensitive patient and donor data.

Challenges we ran into

  • Designing for Inclusion: The biggest challenge was creating a system that didn't exclude people with feature phones. Our solution was the hybrid IVR-app architecture, which ensures universal access.
  • The AI Cold Start: We had to strategize how the platform could be useful from day one, before having enough data for the AI. This led to our phased rollout plan, starting with a rule-based MVP.
  • Meaningful Engagement: Moving beyond simple "gamification" to design a truly respectful and culturally resonant recognition program was a key focus.

Accomplishments that we're proud of

  • The Hybrid Model: We are incredibly proud of the dual IVR and App concept. It's a solution built with empathy, ensuring no one is left behind.
  • Human-Centric AI: Designing an AI that respects donors' time by predicting availability, rather than just sending mass notifications.
  • A Strong Identity: Creating the name and theme of "Raktarpan," which reframes the act of donation as a sacred, selfless offering, deeply honouring the donor.

What we learned

  • For social impact projects, accessibility isn't an optional feature—it's the foundation.
  • The most effective technology is built on a deep understanding of human emotion, trust, and community.
  • A phased approach is critical for building complex systems, allowing for value delivery and data collection long before the final vision is complete.

What's next for RAKTARPAN

Our vision is a phased, sustainable rollout to ensure long-term success.

  • Phase 1 (MVP Launch): Deploy the core rule-based platform (IVR & App) in a pilot region to onboard users and gathering foundational data.
  • Phase 2 (AI Development): Build and validate our predictive models using the data from the pilot, running them in a "shadow mode."
  • Phase 3 (Intelligent Scaling): Integrate the validated AI engine to power smart features and begin expanding RAKTARPAN's operations across new regions.

Built With

Share this project:

Updates