Frantiesco Lucas@frantal97480

VESTRA AI is a hybrid wardrobe assistant that intelligently routes each request across memory, local inference, and remote multimodal AI to optimize latency, token cost, and answer quality in real time. Long Description: VESTRA AI by Frantal Company is a Hybrid AI Wardrobe Agent built to demonstrate intelligent model routing in real-world user interactions. Instead of sending every request to the cloud, VESTRA uses a Hybrid Router that analyzes each prompt and dynamically selects the best execution path: memory lookup, local inference, or remote multimodal reasoning. This reduces unnecessary token consumption, lowers latency, and improves responsiveness while preserving answer quality. The project’s core innovation is not wardrobe cataloging, but routing intelligence and execution transparency. Every request follows a visible pipeline: request intake, intent analysis, complexity estimation, route selection, execution, and response generation. The interface exposes live decision traces, confidence indicators, selected model, estimated cost, and execution timing, allowing judges and users to clearly understand why each route was chosen. VESTRA supports text and image-based prompts, conversation sessions, and a streamlined chat-first experience designed for demos and practical usage. The frontend is built with Flutter Web for a polished cross-platform interface, and the backend is powered by FastAPI with model integrations on Fireworks AI. The architecture is modular and production-oriented, enabling future expansion to additional verticals where hybrid AI orchestration is critical. By combining local/remote balancing, memory-aware retrieval, and transparent AI decisioning, VESTRA showcases how hybrid routing can make AI systems more efficient, affordable, and reliable for end users and businesses.

To participate in the hackathon, simply click Sign up with AMD. If you are not already a member of the AMD AI Developer Program (ADP), you will need to create an account. ADP members can access AI tutorials, experts, and community support. In addition, new members can claim a free month of DeepLearning.ai pro, $100 in AMD GPU cloud credits, and $50 in Fireworks AI API credits. All hackathon participants will receive on hackathon launch additional compute and API credits for their projects. Ready to build what's next on AMD? Sign up with AMD below to secure your spot.