Welcome back to Mind the Skills Gap. In this episode, Stella Collins talks with two of our favourite colleagues, Jan Pannecoeck and Nik Torfs, who are our technical team behind all kinds of clever wizardry at Stellar Labs. We will get into some insights what it's like working at Stellar Labs, but also some of the questions that our customers ask us about the technical aspects of what we do.

‍

Stella: Jan, you've just been promoted to Chief Technical Officer, tell us a bit more about you and your previous experiences, and how this has shaped your role at Stellar Labs (and congratulations on the CTO by the way!)

‍

Jan: Thank you very much! Thank you first of all for getting the chance to become the CTO. Well, I started Stellar Labs almost two years ago. And before that, I did all different kinds of roles. After my studies, as Master in Informatics, I started doing development and then started growing into mobile development. I also worked for the biggest banking companies in Belgium, such as KBC, Belfius and so on.

‍

I started getting into becoming a Scrum Master, because I felt that I loved helping, not only on the technical side, but also improving how people work. And so I became Scrum Master and Agile Coach but I always kept the feeling with the technical side of things.

‍

I really liked doing technical things and even in my spare time, I tried to do some programming and keep up to date with all that's happening in the tech world. And that's how I ended up with Stellar Labs.

‍

I was switching positions and then Raf contacted me, telling me that he was looking for some senior developers who can join and build the platform. And that was perfect timing for me. I took two weeks off first, some vacation, and then I joined. It has been a blast since then. And that was almost almost two years ago, and it has been really fun and time is really flying by.

‍

Stella: And Nik, how about you? How did you arrive at Stellar Labs and where did come from?

‍

Nik: Well, I think it's quite similar to Jan. I started out as a consultant. So I did a lot of different businesses and sectors, different technologies. I think that really helps with what we're doing here right now. I also always had my side business in which I needed to tackle everything technically. So that's kind of a great start to go into the startup. Like with Jan, Raf contacted me at a good moment. I made the choice to get started with Stellar Labs and now we're here for two years already.

‍

Stella: Nik, what aspects of Stellar Labs' mission kind of excites you and resonates most with you personally?

‍

Nik: So for me, a lot of my learning has always come from experimentation and experience. The theory was never really my thing. I could not really study well through theory, but it was always through experience and getting insights that I learned. And so when you, Stella, explained the GEAR model, it really resonated with me. So that got me really excited about joining and being part of this project.

‍

Stella: Jan, how about you?

Jan: For me, I also don’t learn by just reading a book and then somehow you have to know everything. That doesn't really fit with the way I operate. I prefer to do things and just learn by doing. Learning by example is also one of the favourite things that I tend to do.

‍

But also the fact that there is continuous learning happening, we have to continuously keep on learning, it's impossible to just learn until you're 25 and then completely stop learning. The fact that Stellar Labs is there to help companies and to help people within those companies to keep up with learning in a way that it's beneficial for both the company, but also for the people who are doing the learning.

‍

They're learning something that is useful for them, and they can transfer this knowledge into skills. It’s so important for everybody, to continue learning for basically your whole working experience, for lifetime. That was where it really clicked for me. I really love this mission and the fact that we are trying to help people with that.

‍

Stella: And IT it is definitely one of those things where you have to keep learning. It just all the time it just moves along so quickly. So we think a bit about Stellar Labs and the things you're developing there. What are the things that upcoming features, development, sort of the things that you're currently most excited about?

‍

Jan: There's a lot of things happening and as you mentioned, the world is always changing. So with the platform, we are also changing regularly. One of the features that I know is coming very soon is video integration. We already have integration with Synthesia on the platform, but a lot of customers ask that they would like to upload their own videos and want more video integration into the platform.

‍

So, we are almost ready to start building features around video integration. And aside from that one (this is also going to be a challenge and really helpful for our customers too) is SharePoint integration, so that people can start using data files on their own SharePoint and use them together with the Stellar Labs' platform and all the AI features we have already.

‍

Stella: So, right now we can already embed YouTube videos. But I think that further video integration is going to be really useful. And what about you, Nik, what's exciting you right now?

‍

Nik: So I think another one that's really interesting is: Teams integration.

‍

Right now, we're fully a separate platform. But we want to integrate more into the workplace of our customers, right? And we think that for example, with spaced repetition, if we can use Teams to really come into people's daily working environments, in a way that we can really engage them even more with the stuff that they are learning. I think that could be a great extra step forward, to get extra engagements that are needed for learning.

‍

Stella: Yes, and for the work-based actions as well to become really part of your everyday work that's going to be so useful, isn't it? So yes, I'm also really looking forward to that one.

‍

Now, one of the questions that lots of people ask, and I only have a very simple answer for it, so I'm really curious to get and perhaps a slightly more in-depth answer is people are always asking us about where are our servers and I know this relates to people's concerns around GDPR and security.

‍

Jan: First of all, all our servers are of course running in the cloud. We are using AWS services to run everything.

‍

Of course, we are running everything in Europe. And more in particular, I think most of our servers are in Ireland. I actually think all of the servers are in Ireland, and are hosted by Amazon AWS and they are fully ISO 27001 compliant, which basically means that for GDPR, they make sure that all the data is on those servers is not being used outside of Europe.

‍

So we stay within Europe with all our data at all times, and trying to be indeed all the time. GDPR compliant, of course, it's really important that we protect the data of our customers at all times.

‍

Stella: Nik, one of the things that was so exciting in the last year has been the implementation of AI into our platform and I'm really curious about what your thoughts are about how we use it practically to employ it to enhance the customer experience. Perhaps a bit more about the deeper elements of how we use it, can you explain this for us in relatively simple terms, how we're using AI and what's exciting about the use of AI?

‍

Nik: The rise of large language models has been amazing for the kind of development that we're doing. I think there's three or four points where we use AI that are very important to our platform right now.

‍

First, we can generate a great skeleton of the course. So when you want to create the course as a course designer, you put in three or four inputs and we will generate transfer behaviours based on that and from those transfer behaviours, we will generate a full course for you.

‍

Of course, you'll still have to check the course, change some details, but it avoids that white page syndrome when you don’t know where to start when you're building a course. So, it's a great kickstart that's how I typically look at it.

‍

But then apart from that, when you start adding content yourselves or new modules, we can generate scenarios for you, which are basically the small kind of role-playing scenarios, stories where you can apply your knowledge in a safe way. So the AI generates a situation based on the content of your course. And you will have to choose a certain direction you want to go and choose how will you deal with that situation. But of course, as it's all a role-playing game, you don't have to fear about making mistakes. So it's a really great way to first apply your knowledge that you've learned then after that we have work-based actions.

‍

Of course, a designer can create their own work-based actions in the course. In this case, AI will suggest some good actions, based on the course.

‍

And then I think lastly, we do quiz questions and assessments at the end of every module in the course in the GEAR model. And those we can generate as well, such as fill in the blank questions, multiple choice, the platform gives you some great suggestions. You can always make your own but it always gives you a good start and some ideas using the content of the course.

‍

Stella: Yes, sometimes the AI comes up with things that maybe you wouldn't have thought of. Which again, you can you can choose to use or you can choose to discard but it just saves so much time: certainly the spaced repetition questions, it's so boring to think of questions about what people are learning. So that's really useful.

‍

We're also very lucky that we have the full support of Donald Clark, who's probably one of the world's experts right now on AI and learning. And he fully supports what we're doing and has been incredibly helpful with contributing ideas of how to use AI. So I'm sure there's an incredible future for AI and I think we're going to be using it in more different ways apart from just generating content in the future, but that is for another day perhaps.

‍

Nik, for our listeners that aren't as tech-savvy as you, I often hear people talking about front-end, back-end. And I have a kind of a vague idea of what it means. But I think it would be useful for people to understand what's the difference between the front end and the back end. And how does that work?

‍

Nik: So the front end, I always explain as the piece of the software that you interact with as a user. So you go to your browser you go to the Stellar Labs platform and what you see there, that is the front end of the application. It is basically all the code that runs in the browser.

‍

The back end is typically the bulk of the business logic. Like a lot of that is happening there, but the user typically doesn't see anything of that.

‍

So a lot of the times in the conversations we have, we say “There's a lot of back-end work”. That's typically not what the user sees or what makes a big impact, because they cannot use it until there is a front-end and that makes use of all of that work.

‍

Stella: Okay, that's really interesting. And then, how does UX kind of fit into that that presumably is like definitely as a front-end thing, I guess.

‍

Nik: Absolutely. So UX means user experience. So it's about having small animations, but also the kind of paths that a user has to take to get to a certain point.

‍

Is it all intuitive? Is it all clear? Can they navigate the user interface without needing to read the manual? That's all super important in the age of web applications.

‍

Stella: I guess that's also where the accessibility piece is particularly important or does that also dive into back-end?

‍

Nik: No, accessibility is indeed also a front-end thing. There we focus more on getting some keyboard access as well for people that maybe cannot use a mouse. That's a very, very broad area to go into as well.

‍

Stella: We've definitely could get an expert to come and talk about accessibility. I'm always curious about it because it's all about the end user in the end. Okay, so that helps to understand what the difference between back end and front end.

‍

What about the technologies and the kind of the programs we use at Stellar Labs now? You know, I was a COBOL program many years ago. And I know for certain we're not using COBOL. But what do we use?

‍‍

Jan: I'm really glad that I can confirm that we're not using COBOL anymore. I learned that in school. I never used it afterward. As Nik already explained, we have a complex back-end. For the front end, we typically use TypeScript, which is basically a kind of JavaScript, but safer, in combination with React, and that's what we are using on the front end.

‍

On the back end, we are running partially on Java, and also partially on Node.js. We are building basically a serverless system on AWS, as we mentioned earlier, where we, based on events, try to communicate between all the different microservices that we have on our back end. The front end is quite a complicated system that's running over that. And then of course, as mentioned before, we use AI quite a lot and large language models.

‍

Stella: And presumably, we are using all those different languages because each one kind of serves a particular purpose and expertise, and that's why we use different ones. Is that right?

‍

Jan: Yes, indeed. We're trying to do in Node.js all the time because it just faster is easier to implement also. And the Java is because the biggest part of our backend was written or already in Java. So we kept on maintaining that part. We kept on adding new functionalities over there, of course, but each language has its own reasons why we are using it. And based on our own experience from the past and our own knowledge, we also have to keep that in mind.

‍

Of course, we don't want to add too many new things all the time because new languages are appearing regularly. But we also have to make sure that it's maintainable for us also, to keep on working everything. And I think that's something that sometimes people don't necessarily think about the maintenance of code is just as important and the kind of the recording and the communicating about code, so that if you have to go back in and look at somebody else's code, you can understand why that code was written and kind of how to fix it.

‍

And we all have to understand each other's code all the time. And we can't just keep on inventing new things, or starting new tools or new platforms all the time.

‍

We have to stick to something because we also trust the platform. Of course, we have to make sure that everything keeps on running stable and robust has been mentioned before.

‍

Stella: If we find a bug or if somebody reports a bug, you are really quick at A) finding that bug and B) be fixing that bug. What is it that helps you do that?

‍

Jan: Well, first of all, as mentioned, we both are working with Stellar for almost two years. So both of us know the code by heart. Almost every line of code we have touched at least once and then we also have a system in place which is basically monitoring all our services. And if something goes wrong, it notifies us and it gives us a reason why something goes wrong.

‍

So if some exceptions are happening, we can easily trace what's happening, why is it happening and then we can fix the problem quickly. So that's definitely also help with the fact that this system is in place and we can easily trace failures in the system.

‍

Stella: Ok, because I’m always very impressed with how quickly you you solve anything that happens. And we all know that there is no coding without the occasional bug.

‍

So, as a final question, we want Stellar Labs to be robust. We want it to be safe. We want it to be reliable. What are the things that really help us do that?

‍

Jan: So, I think one very important factor is that we built the platform on AWS, what really allows us to scale the platform and to also just maintain a great uptime. We're not managing any servers ourselves. We just get that as part of our contracts with Amazon Web Services. And that helps a lot because managing servers is a whole different kind of work, that takes a lot of time. And if we can avoid that, that just helps tremendously. So we can let the experts in AWS do their job. If we need to grow or when we get a lot of new customers, we don't have an issue that we need to get new servers because it just automatically scales up. So it's perfect for the kind of business that we are where we are in the growth scenario.

‍

Stella: I know we have various kinds of collaborative projects going on where when we do what we do best and then we can work with other people who do what they do best. Anything else that you know, it's really important in terms of us being robust and secure.

‍

Jan: In terms of security, we ask an external company to regularly do a security test, of course, to make sure that our platform is secure. And those are really helpful for us and we also learn out of those reports each time.

‍

And last time we did one earlier this year, we got a very good result, so we were really happy about that. We did quite a lot of changes in the past six months based on security. And the last report confirmed that indeed all the changes we did were very good and really helped us on improving our security and having them really at the top level right now.

‍

Stella: This is a kind of random question, but we usually ask it at the end of this mind the skills gap podcast, what are your tips for anybody who wants to learn a new skill? Jan, do you want to go first?

‍

Jan: Just get started. The first time you do something you probably fail. That's not a problem. And the second time might also be a hard one. But if you can keep on and you managed to do something 10 times, 20 times, or 100 times eventually you will get there.

‍

So, just get started and try to not give up and keep on the spaced repetition for example, keep on repeating what you've learned and get it into your long-term memory instead of the short term.

‍

Stella: And that's really very important. What I really liked as I was watching you say that, was you were using your hands and I think what you really wanted to say, is “get it in your fingers” and I think for skills is really important that it is not just our brains that we need to get information and skills in, but also get it into our bodies and fingers. What about you, Nik?

‍

Nik: I very much agree with what Jan said there. For example, what I do for example when programming something new, I will typically just start with first versions of something, you just do it to learn from it. Then you throw it all away and you start again and then from what you learned, you will do it better.

‍

So, if you're learning something, start doing it so you can learn from doing it. Of course, any kind of extra information you can get theory or anything else is great material to start from, but without doing it I don't think you can ever really learn something.

‍

Stella: I think you're really right. And I like what you I was kind of implying from what you were saying is: try it on something first and don't be precious about that first attempt. If it's wrong or it doesn't work or if it looks ugly, that doesn't really matter. Just you can just throw it away and think “okay, now I can I can do it again.”

‍

Nik: yeah, that's a great explanation of what I was trying to say. Thank you.

‍

Stella: You said it beautifully yourself.

‍

Jan and Nik, thank you both so much. I really enjoyed this conversation with you. I've understood a bit more about you know, the kind of the back end of Stellar Labs, the bit that perhaps the public doesn't see. And I want to thank you both so much for being on the podcast and being such aced developers within Stellar Labs so thank you both, and see you soon.

‍

Nik: Thank you for having us.

‍

Jan: Thank you, Stella. See you soon.

‍

You can listen to the full episode here: