TFT: TeamFood Tactics

Where AI automatically finds you food

Comment

Inspiration

"Americans waste about 60 million tons of food every year."

Food insecurity and food waste are major challenges in our communities. As we saw local bakeries and food banks struggling with leftover food and limited resources, we were inspired to create a solution that not only reduces waste but also bridges the gap between surplus food and those in need. The idea was born from the desire to make sustainable food distribution accessible—even for those without smart devices—by leveraging a simple, voice-first interface.

What it does

TeamFoodTactics is an AI-powered sustainable food distribution platform that connects suppliers (like gluten-free bakeries with surplus bread) with users who need specialized diets (such as individuals needing gluten-free options). Users simply call a dedicated phone number to describe their dietary requirements or inventory details, and our system:

  • For Suppliers: Facilitates the donation of leftover food by allowing bakery owners to update available food through voice calls or image uploads.
  • For Users: Provides personalized recommendations for the nearest food source that meets their specific dietary needs.
  • Behind the Scenes: Uses advanced AI to process voice and image data, generate dynamic SQL queries, and match supply with demand in real-time.

How we built it

System Architecture

  • Frontend: Next JS, Tailwind CSS
  • Backend: Python, Fast API
  • API: Retell AI, Open AI, Open AI Swarm
  • Database: Supabase

Challenges we ran into

  • Voice Interface Integration: Ensuring a natural, seamless voice interaction over any phone, particularly for users without access to smartphones.
  • Dynamic Query Generation: Training AI models to generate accurate SQL queries based on variable user inputs was complex and required extensive testing.
  • Real-Time Data Processing: Managing and updating live inventory data from multiple suppliers, including the OCR extraction from uploaded images, presented integration challenges.
  • Scalability: Building a system that is both secure and scalable to handle the unpredictable nature of food donation and distribution volumes.

Accomplishments that we're proud of

  • Inclusive Design: Successfully implemented a voice-first interface that serves digitally underserved populations, making it accessible for anyone with a phone.
  • AI-Driven Matching: Developed an innovative system where AI dynamically generates queries to connect surplus food with those in need in real time.
  • Seamless Multi-Component Integration: Efficiently combined telephony, OCR, geolocation, and real-time database management into a single, robust platform.
  • Sustainability Impact: Created a tangible solution that not only reduces food waste but also improves access to nutritious food, driving positive social and environmental outcomes.

What we learned

  • Interdisciplinary Integration: The project deepened our understanding of integrating conversational AI and other technologies.
  • Agile Problem Solving: We learned to tackle challenges iteratively, refining our approach to dynamic query generation and real-time data processing based on user feedback.
  • Scalability & Security: Working with Supabase and integrating various APIs helped us enhance our skills in building secure, scalable systems that can adapt to growing user needs.

What's next for TeamFoodTactics

  • Expand Language Support: Incorporate additional languages and dialects to serve a broader, more diverse user base.
  • Mobile App Integration: Develop a companion mobile app to complement the voice-first interface, offering enhanced functionality for tech-savvy users.
  • Predictive Analytics: Implement machine learning models to forecast inventory needs and optimize logistics for food distribution.
  • Enhanced Personalization: Introduce user profiles, history tracking, and dashboards to provide personalized recommendations and sustainability metrics.
  • Partnership Expansion: Collaborate with more food banks, community organizations, and local governments to scale the impact regionally and nationally.

Built With

  • fastapi
  • nextjs
  • openai
  • openai-swarm
  • python
  • retellai
Share this project:

Updates