BrightNight recap - Leveraging AI and data virtualization for QA in ERP implementations

October 12, 2023

Joke Gijsbrechts

Hi, everyone! We want to take a moment to share our thoughts and insights from the recent BrightNight session. We delved into the world of software testing in the era of AI. As individuals who have spent years as quality and test managers, we couldn’t help but be intrigued by, and also doubt, the potential impact of AI in our field.

The focal point during this session was to explore how AI can be used to enhance the quality aspects of ERP implementations or migrations. Moving to a new ERP system is a challenging task, especially when it comes to ensuring the utmost quality. ERP implementations involve a multitude of business scenarios, each tailored to meet the specific requirements of our clients. And let’s not forget the differences between various ERP packages themselves.

 

BrightSQAI

Enter Dean Bodart from Brightest, who saw an opportunity and transformed it into a practical approach for our ERP clients. Dean managed to leverage the power of AI to automate the generation of contextual test cases on a massive scale. The technical process behind this achievement involved vector databases and structured prompts fed into BrightSQAI, Brightest’s very own ChatGPT version. This process also incorporated historical data from ERP systems, resulting in a well-organized set of structured test cases.

To put it into perspective, BrightSQAI accomplished in 48 hours what would take an experienced tester a staggering 62 Man-Days to achieve. The ROI of leveraging AI to determine test scenarios in an ERP environment is nothing short of impressive.

As the icing on the cake, these test cases are delivered in a user-friendly Excel format. Next to that, they can be seamlessly integrated into the most popular test management systems like SAP Cloud ALM and Azure DevOps.

 

Data virtualization

In addition to generating a library of test cases, Dean also delved into the realm of test data. We all know the pain associated with this aspect of testing. Dean had another Bright idea here, taking us on a journey from synthetic data to semi-synthetic data. And ultimately leading to a virtualized data set. This means that we can store massive amounts of data in a virtualized database, which offers substantial advantages. Delphix, a company that provides this service and was also present at the event, proved to be a valuable partner in this endeavor.

 

Conclusion

Still, Dean concluded with an important reminder. While AI and data virtualization are powerful tools, the expertise of experienced test specialists — such as functional test engineers, automation engineers, and QA managers — remains indispensable for project success. We’re not quite at the point where we can simply push a button and deliver everything to our clients without any human intervention.

Nevertheless let’s be honest, the use of BrightSQAI and data virtualization in ERP implementations offer undeniable benefits. They represent a significant leap forward in a time when the pressure to accelerate time-to-market is growing. And the consequences of a lack of quality can be substantial. This BrightNight illuminated the path toward a more efficient and effective future in ERP quality assurance, where AI and data virtualization play pivotal roles in achieving success!

Co-written by Catherine Ooms & Joke Gijsbrechts

BrightNight recap security

BrightNight recap - Cybersecurity

During this session, Stijn Jans from Intigriti talks about how we are all hackers. He also discusses the difference between penetration testing and crowd hacking.

Read the article