← Back to Ideas

A Robot Sous Chef

April 20, 2022

I often need quick answers to my culinary questions. Why should I need to scroll through a whole blog post?

The Spark Notes Version


When we're cooking, there are two types of aids that we need. The first is recipes, or long-form help. This is a pretty solved problem.

The second are short answers to straightforward questions. How long should I boil an egg to get a perfect soft boil? Can I substitute margarine for butter?

The problems of getting answers to these question is not yet solved.

This would be an app or website where you can ask these questions. Our AI would then scrape the internet, and all of the recipes we've catalogued, and give you our best answer. We'd also provide references so you can make sure that our answer is accurate.

Ideas like this, right in your inbox.

Let's be honest. You don't want to have to come to our site every time you want a new idea. Let us send ideas like this one right to your email, for free, every week.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

The Long Version


Search engines have a problem. They can't let articles with only 10 words rank, because otherwise all search results would get even more cluttered with spam than they already are.

But sometimes, when you have pasta dough flung over your shoulder and hands covered in garlic, you don't want to have to scroll through someone's life story to figure out if there's a good substitute for coriander.

This would be an app that, at the core level, simply synthesizes all publicly-available cooking materials.

Then, our AI would be able to answer any question you have succinctly and accurately.

If you wanted to ask it how many times you should fold croissant dough, it would just need to go through the croissant recipes we have and figure out what would be recommended.

This could most likely be done with a pre-made algorithm such as GPT-3. If you simply train it using cooking data, it is very likely that you could get accurate results without needing to make your own AI.

Of course, you could also do this without needing AI. If you used Google to find the top 100,000 culinary questions being asked, you could simply compile a database of questions and answers, and just use AI to match different phrases of the same question to the appropriate answer.

Kind of like stackoverflow, but for food.

For monetization, you could go a bunch of different ways. You could make this a paid service. That would decrease your user base, but might be the most efficient method of collecting revenue. Of course, you could also simply use display ads. Since your customers would likely engage in significant word of mouth, your daily users should only increase over time.

This would simply be an immensely useful tool, and if you make it, I'd love to be your first customer.

If you liked this idea, subscribe to our newsletter! We'll send you one great business idea like this every week.