.png&w=256&q=75)
🤓 Latest Submissions
FAUNA - Field Alert Using Nature AI
Problem being solved: Every year millions of people encounter dangerous wildlife with no idea what to do. FAUNA gives anyone a 10-second species identification from a photo, immediate recommended actions, specific first aid guidance if bitten, and a one-tap emergency message with GPS coordinates. Built for hikers, farmers, safari guides, and anyone working in environments where dangerous wildlife encounters can be life-threatening. Target user: Hikers, outdoor workers, farmers, safari guides, and tourists in regions with dangerous wildlife — snakes, spiders, scorpions, crocodiles, venomous plants, and jellyfish across Australia, Africa, South and Southeast Asia, and Southern Europe. How native.builder was used: The entire application was built using native.builder during the hackathon period. native.builder's AI agents generated the application structure, React component architecture, three Supabase Edge Functions (wildlife identification, nearby emergency services, emergency message generation), and the full mobile-responsive UI. External integrations added: Claude Vision API (Anthropic) for species identification, Foursquare Places API for nearby hospital and pharmacy search, Web Share API for emergency message distribution. All deployment, secrets management, and Edge Function configuration was handled through native.builder.
10 Aug 2026
.png&w=828&q=75)
Hybrid Semantic Routing Agent
Large Language Models provide high-quality responses, but sending every request to a remote API increases latency, cost, and token consumption. This project explores a hybrid routing architecture that minimizes remote inference while maintaining response quality. The application combines three complementary strategies. First, every incoming query is converted into a sentence embedding using Sentence Transformers. These embeddings are indexed with FAISS to create a semantic cache capable of retrieving previously answered questions even when they are phrased differently. When a semantically similar query is found, the cached response is returned immediately, eliminating unnecessary inference and reducing response time. If no suitable cached response exists, the routing engine evaluates the complexity of the request using lightweight heuristics based on query length and keywords. Simple factual questions are handled by a lightweight local language model, while more complex reasoning tasks are routed to Fireworks AI. If the remote service is unavailable, the system automatically falls back to the local model, ensuring robustness and uninterrupted service. The project records cache statistics, estimated token savings, routing decisions, similarity scores, latency, and model selection for every request. Its modular architecture allows additional routing policies, multiple remote models, or persistent storage such as Redis or SQLite to be integrated with minimal changes. The primary objective is to demonstrate that intelligent routing combined with semantic caching can significantly reduce remote LLM usage while maintaining a simple, extensible, and production-oriented architecture.
13 Jul 2026
