AI and Machine Learning in Mobile App Testing

Getting great apps to the market as early as possible is the key to business success. But the fact is that there are so many things needed to be done by mobile developers in this process. Bitbar Cloud is an all-in-one mobile DevOps platform that aims at easing the daily work of all mobile app developers with unique flavors in DevOps tools and services. And we want to make building and testing applications easier.

The DevOps movement says “shift left” which really means test early and test often. This is great but it puts a lot of responsibilities for testers and DevOps teams. For many mobile application teams, they might just have not enough resources to have a dedicated mobile app test engineer (or it is hard to find any) that would write all the required automated tests. Even larger organizations want their testing teams to spend more time creating automated tests for the planned and must-work user scenarios. But can any team really afford the time to do exploratory testing manually for each new release or build? If no, how should you still handle the need of ‘shift left’ or ‘test early and test often’?

Download this manifesto to learn what it takes to embrace mobile DevOps today

Today’s mobile teams have a great selection of mobile application testing frameworks to help create functional test cases, but what we are missing are tools making the creation of test suites easier and faster. We need solutions that will know how to execute smart tests on mobile applications against real devices with no need to manually show how and maintain the once created tests. We all still remember those record and replay tools for testing from some decades ago…

Luckily there is a new computer science field that is addressing exactly this problem. Machine learning (ML) and Artificial Intelligence (AI) are what many predict the solvers of this problem. There are two ways to tackle this, either you create your own specific AI that will know how to test your application in the best possible way or, you take an existing AI general solution that will do advanced exploratory testing on all kinds of applications, including mobile games.

We have decided to follow the second approach as we have thousands of different applications being tested in our device cloud regularly. Creating a good general AI-powered testing method allows us to provide a solution that helps most of our customers to ease the smoke testing of their applications. This exploratory testing will uncover any random misbehavior without any of your engineering time.

We are thrilled to roll out Bitbar’s AI Testbot, our AI-powered test automation solution within our mobile device cloud to help you automate app testing easier, faster and smarter. It will be available later week or early next week to all of our paying users including Bitbar Enterprise customers with any type of environments and installations. It is first available for Android devices (Android OS 5.0 and above) and later for iOS devices. (For information about getting access to trying out this new functionality, please get in touch with our sales at sales (at) bitbar (dot) com).

Below is a short video recording of one sample AI Testbot tests. In the next blog (coming soon), we’ll provide more details on this new feature on Bitbar Device Cloud and how it compares to its counterpart AppCrawler.

For technical questions and feasibility questions, please get in touch with me at niko (dot) cankar (at) bitbar (dot) com.

The Must-Test Global Devices for Mobile Device Testing

Get a sheet of the most popular global smartphones and learn what devices every app developer should test against and verify the compatibility.


  • ondemand QA

    We need some more articles on web app testing.

  • Absolutely agree, ‘We need solutions that will know how to execute smart tests on mobile applications against real
    devices with no need to manually show how and maintain the once created tests.’ The application surface is getting complex each day and various applications interact with each other through APIs, which add up to the complexity. Apart from increasing the complexity, there is immense rush to bring the application to the market. With demanding market scenario, the releases that would happen over a month’s span are being done within a week’s time. This is putting tremendous load on testing. Hence, Machine-based intelligence is needed to overcome the testing and QA challenges that testers would face on a regular and recurring basis.
    On similar lines, you might like to check out this post ‘Can Artificial Intelligence power up your App Testing efforts?’.. here’s the link for your reference –

  • Asiq Ahamed

    Does Bitbar support Real User Condition Testing as mentioned in the below link?