February 23, 2017

Clarifai Featured Hack: Make The Perfect Toast With Toastifai, An AI-powered Toaster

Table of Contents:

Toastifai is an AI-powered toaster that uses Clarifai’s visual recognition API to produce the perfect piece of toast every time. Using Clarifai’s Custom Training, the Toastifai team trained a visual recognition model on photos of toast in different stages of toastiness and used the results to build “smart” toaster!

In honor of National Toast Day, we have a solution for all your toast-burning woes! Toastifai is a “smart” toaster that can help you reach toast nirvana every single time. Using Clarifai’s Custom Training, Toastifai can recognize toast in various states of toastiness and pop out toast at exactly the moment it reaches toasty perfection – and send you a text alert that it’s ready. Say that fast ten times!





Obviously, Toastifai is the greatest thing since sliced bread! We love when hackers combine our software with hardware – particularly hardware that can create delicious edible enjoyment. Check out Toastifai’s GitHub repo here!


We caught up with Jessie Pullaro and Frank Callas, students at Florida Polytechnic University, to talk about Toastifai.

Clarifai: What inspired your idea for Toastifai?

Jessie: My group and I all like toast. We all wanted something to make perfect toast and we made Toastifai!

Frank: We were in the car talking with our friend, Muneer. Now Muneer is a busy guy. He was telling us that he hates making toast because he either burns it or he leaves it because he gets distracted and doesn’t remember to get it before he leaves. What we set out to do on October 15th, 2016 was fix Muneer’s problem.

How did you build the app?

Jessie: The entire project is in Python. We also used openCV and a package call Pillow. We also used Twillio for the texting part. Oh! And the hardware was a Logitech camera and a Raspberry Pi with an Arduino and temperature sensor to tell the webcam to turn on.

Frank: To take images from a webcam we used openCV and to manipulate it we used pillow, which uses numpy. We also used Clarifai to determine when the toast was ready, which also included using the JSON library. All of this was on the Raspberry Pi.

What was the best part about working with the Clarifai API?

Jessie: Honestly, it was the easiest thing in our project to use.

Frank: It was SO EASY TO USE! The web app for it has DRAG AND DROP! HOW MUCH BETTER DOES IT GET! Plus there is super low latency, which was great since we had to do real time processing.

Thanks for sharing, Toastifai team!

To learn more, check out our documentation and sign-up for a free Clarifai account to start using our API – all it takes is three lines of code to get up and running! We’re super excited to share all the cool things built by our developer community, so don’t forget to tweet @Clarifai to show us your apps.

And give the Toastifai team some props in the comments below. Until next time!