Participated in public demo 2505 with AI Nutritionist

The problem [00:25] A global shift toward healthy living, plus millions of people managing food allergies or specific diets like keto and vegan, has made reading a food label or restaurant menu harder than it should be, the information is often complex, misleading, or incomplete, and existing apps only offer generic scores that don't answer the real question: is this right for me, specifically, given my own conditions and diet. What they built [01:40] FoodScanalyzer is a personal AI nutrition advisor: users build a one-time health profile (allergies, conditions, diet, taste preferences, symptoms), then scan a product barcode or label, or photograph a restaurant menu in any language, and GPT-4o Vision reads it while a set of specialized AI agents, a scoring engine, a risk-flagging agent, and a coaching layer, tied together by an n8n orchestration layer, turn it into a personal fit assessment (suitable, caution, or avoid) against that specific profile, rather than a generic nutrition score. What happened [05:35] An evaluation across 5 user profiles with varying restrictions against 3 common food items, scored on flag accuracy, score accuracy, and clarity, showed the system correctly flagging what to avoid for restricted profiles while correctly giving unrestricted profiles a green light, avoiding false positives and validating that personalized, per-profile advice is achievable. The ask [07:05] Jack is looking for strategic partners (founders and retail/health-tech leaders) to help shape go-to-market strategy, experts in AI explainability and nutrition to validate the AI's recommendations, and early-adopter "pioneers" to join the private beta and give feedback.