Bettermeals

• 

  onboarding

• 

  meal

• 

  grocery

• 

  cook notifications

Inspiration We were inspired by a simple question: “What if your meals could be chosen by both your biology and your taste?” Most meal plans ignore your labs. Most health food ignores your cravings. We wanted to fix that -so we built BetterMeals.

What it does BetterMeals takes your lab results, food preferences, and lifestyle inputs, then auto-generates a personalized grocery list and 3-dish cook plan, then message the cooks the plan. We use Qloo’s taste graph to ensure each meal not only meets your health needs, but feels familiar, exciting, and satisfying.

How we built it

• Lab Parser (custom): Extracts biomarkers from PDF/txt lab reports
• Profile Engine: Assigns you to a nutrition plan based on goals like inflammation, sleep, muscle repair, etc.
• Nutrient Target Generator: Converts your profile into daily micro/macro targets
• Meal Generator: Selects dishes that hit those targets in ≤45 minutes and ≤3 dishes
• Qloo Integration: During onboarding, Qloo refines dish selection to match user taste- e.g., if you like “Mediterranean + tofu,” Qloo might elevate a tahini-lentil bowl
• Grocery Engine: Merges all ingredients into a smart shopping list (with future e-commerce integration)
• Cook Console: Auto-generates daily step-timed recipe cards with plating instructions, voice-assisted walkthroughs, and WhatsApp push - so your household cook can execute the plan without needing to “figure anything out”

Challenges we ran into

• Parsing lab reports reliably across formats
• Balancing nutrient constraints with taste.
• Building a household-level planner that satisfies multiple profiles
• Making healthy food feel joyful, not clinical

What we learned

• Taste compliance is as important as nutrient compliance, Qloo made that real :)
• Even evidence-based nutrition needs UX-level storytelling
• Household-level meal planning is a deeply complex but solvable optimization problem
• Building in modular layers (labs → profiles → taste → meals) is critical for scaling

Built With

• and-lab-parser-pandas-/-numpy-?-nutrient-math
• and-profile-modeling-openai-gpt-4-/-langchain-?-natural-language-lab-parsing
• bigbasket)-firebase-?-user-auth
• fastapi
• figma
• firebase
• google-sheets
• grocery-list-view
• ingredient-merging
• langchain
• meal-descriptions
• numpy
• nutrient-engine
• onboarding-flow
• openai-gpt-4
• pandas
• playwright
• python
• qloo-api
• real-time-cook-updates-whatsapp-cloud-api-?-daily-recipe-playbook-and-reminders-to-household-cook-figma-?-ui-design-for-resident-onboarding
• recipe-generation-qloo-api-?-taste-intelligence-layer-for-personalized-meal-preferences-fastapi-?-backend-api-for-modular-meal-generation-playwright-(python)-?-browser-automation-for-grocery-e-commerce-flow-(e.g.
• render
• vercel
• whatsapp-cloud-api