What We Do

Matelab makes it easy for machine learning novices to use machine learning in their applications. Using a simple step by step interface, users can create and train a model which can easily be implemented into their program using a few API calls. In short, machine learning for all!

We are still in a closed beta, but if you are interested in trying out our service click the link below.

How it works

We want more people to learn, and use machine learning in their day to day lives, so we designed a simple step by step UI to guide you through the process. Getting started is simple. We guide you from naming your model to selecting the data to train it with.


We will ask you all the questions and you tell us what you want. It's as easy as filling out a form. Once your model is built, training it is just as simple.


Once your model is built and trained, we allow you to test it via our web interface. Over time, you can continue to train and test it to make it more accurate and robust.


Once your model is tested, integrating it into your application is as simple as making a few API calls. Changes to your model will automatically be updated in your application.


Applications for machine learning are vast, but we thought it would be good to highlight some of the big use cases.

Object Detection

One of the simplest and most useful models you can build is for object detection. Have you application detect and notify users when a specific object is seen!

Facial Recognition

Create a model to quickly identify specific people. Some notable applications for this type of model are security or tagging photo albums.

Intent Detection

When building applications such as chatbots, understanding the request is key to it's success. Creating a model to look for similarities in sentences is a good start to solving this problem.


We support many different types of data to train your models with so that you can truly customize it and get what you need. We are also dedicated to integrating with other services such as Twitter so that you can easily pull in large amounts of training data.





Featured In