This month I had the opportunity to meet in person Alyvix’ founder. Therefore I thought… Why not asking him some questions about Ayvix’ technological change? So here we are 🙂
Can you briefly summarize yourself?
Hi everyone, I’m Alan Pipitone and I’m 33 years old. In 2014, when I moved to Lugano, I decided to create the Violet Atom, which has the primary purpose of providing consultancy applied to the world of robotics and automation. If I should define myself, I’d use the word “researcher”, because I love keeping myself up-to-date. I love reading about new technologies, attending courses on programming languages, robotics, electronics, Artificial Intelligence, Machine Learning.
How did you get the idea that led you to “create” Alyvix?
Back then, I was working as a system engineer in an oil company. One of my main activities was to verify the correct functioning of the business-critical applications (CAD, ERP systems, etc.). We used to have Nagios and other monitoring systems to check servers’ status (CPU, RAM, and so on).
Those checks and monitoring activities weren’t enough, though.
Employees often occurred in error pages, even if checks didn’t find any anomaly. In that time, I was approaching and particularly fascinated by computer vision. Therefore, I had an idea: simulating the human eyesight – that is creating a software able to see what is happening on the computer screen and to interact with it and to send alarms when what we expect to appear on the screen doesn’t.
A while back, I attended an IT conference in Bolzano, where I met Georg Kostner, System Integration Business Unit Manager at Würth Phoenix. I decided to write him an email to talk about my project. This was the beginning of Alyvix.
How would you describe Alyvix’ technological change during these years?
Well, I think that its technological growth is definitely positive.
In the beginning, Alyvix was written in C++. It didn’t exist a frontend and test cases were executed thanks to XML files duly compiled. Subsequently, I tried other programming languages. In the end, I chose to use Python for its ease of use and its great potential. I modify an opensource editor to let users write test cases directly in Python, offering utilities in order to automatically create code snippets that used Alyvix’ features. Then, I added to Alyvix’ core an OCR (Optical Character Recognition), which recognized text from screenshots captured by Alyvix.
Now Alyvix uses a new editor, which allows users building test cases simply interacting with the computer screen and compiling a table with the right logic flow. Moreover, it has been improved also the measurement system, which is more precise and needs fewer hardware resources than ever.
This is a simple example of the entire test case creation process:
But…. There is still a lot to do!
In fact, future releases are going to be particularly interesting and disruptive. Alyvix’ roadmap envisages lots of important changes, which you are going to discover if you stay tuned 😉
The most important news is that Alyvix can be used as a standalone tool by the end of summer (at least the first version). This means that you won’t need a monitoring system to schedule Alyvix test cases anymore.
Is Alyvix a scalable system able to face future challenges?
Yes. We are working on new machine learning algorithms to improve Alyvix’ core and, as said before, on a new editor to make easier to write a test case. As for scalability, we are working on a system to distribute and automatically synchronize test cases on more probes. Therefore, it’s not a problem to manage lots of probes and test cases.