got dinner

AI powered recipe discovery from your own kitchen.

problem

I have always loved cooking. When I was younger, I remember wanting to have that skill of opening a fridge, looking at what was there, and somehow knowing what I could make. It felt like a kind of everyday creativity that I really admired. As I got older and started cooking more, I realised there was another side to it. I would sometimes buy ingredients for a recipe, use part of them, and then forget about the rest. Things would slowly disappear into the back of the fridge until they eventually had to be thrown away. I realised that the problem was not that I needed more recipes. I needed a better way to make the most of what I already had. That became the starting point for Got Dinner.

solution

I created Got Dinner, an AI cooking assistant that helps you turn the ingredients you already have into something worth cooking. Instead of starting with a recipe and buying everything around it, Got Dinner starts with your reality. You take pictures of your fridge, pantry, and freezer. AI analyses what you have and uses those ingredients to create a recipe around them. The idea is simple: Don’t ask “What recipe should I make?” Ask “What can I make with what I already have?” I wanted to combine something I genuinely love, cooking, with something I became increasingly conscious of as an adult, food waste.

Got Dinner is quite personal to me.

When I was younger, I wanted to become the kind of person who could look at a random collection of ingredients and create something good from it. I loved the idea that cooking could be intuitive, creative, and resourceful.

As I became an adult and started buying and cooking my own food, that idea took on a different meaning. I became much more aware of how easy it is to waste food. You buy something with good intentions, use half of it, and suddenly weeks later it is sitting forgotten in the back of the fridge.

I started thinking about whether AI could help solve that in a way that felt natural.

What if, instead of searching for recipes and then buying ingredients, I could simply show an AI what I already had and let it figure out what was possible?

That question became Got Dinner.

I built Got Dinner from the ground up as a working AI product, using AI as part of the development process itself.

The product is powered by Google Gemini, which handles the multimodal understanding behind the experience. Users can provide images of their fridge, pantry, and freezer, and Gemini interprets those images to identify ingredients and generate recipes around them.

The application is hosted on Vercel, giving me a fast and scalable environment to deploy and iterate on the product.

For development, I worked alongside ChatGPT and Claude throughout the build as a development partner to explore architecture, prototype interactions, debug problems, refine prompts, and iterate on the product.

The project became an experiment in what it looks like to build a product in an AI native way: combining my own product thinking, design decisions, and development with AI systems that could help me move from an idea to a functioning experience much faster.

Hosting: Vercel

Core capability: Multimodal image understanding + AI recipe generation

year

2026

timeframe

4 months

tools

HTML, CSS, JavaScript, AI: Google Gemini, development on ChatGPT + Claude