EASEY PREP

Inspiration

What it does

It allows users to plan out a week's worth of meals, with priority given to whatever is currently in your fridge.

How we built it

We built the front-end using Vue, created a basic backend using Supabase, and selected Groq to be our interface between the LLM and the project.

Challenges we ran into

We ran into some issues involving token limits while prompting the LLM, alongside hallucinations related to meal statistics. We ended up solving a lot of these problems by implementing logic on the front end to fill in the gaps where the LLM tended to fail.

Accomplishments that we're proud of

We were able to get a very good looking Abstract Brutalist design for our front-end.

What we learned

Never use Google Gemini.

What's next for Meal planner

Creating a custom AI model trained on hundreds of different recipes to create more accurate results than just prompting a single LLM.

Built With

• supabase
• tailwindcss
• typescript
• vue