Meet Iulian, a former intern on the CryptPad team whose passion for cybersecurity and cryptography runs deep. His bachelor’s thesis aligned perfectly with his internship, making the experience at XWiki both relevant and eye-opening. For the first time, he had the chance to put cryptography into practice in a real development environment.
Iulian helped pave the road toward quantum-resilient cryptography inside of CryptPad, as well as the ways to make the transition possible. This was done by being fully integrated in the team and getting familiar with the CryptPad codebase, especially its cryptographic stack. Besides cryptography, he showed great interest in FOSS and real-world privacy and ethical concerns.
Let's discover below the challenges he faced, his thoughts on the team spirit, and his impressions of the open-source world.
- Favorite artist: Set It Off. I am a rocker in my spare time.
- Best book ever read: "Darksiders: The Abomination Vault". I am a sucker for fantasy works.
- Dream destination: Switzerland
- Quote you live by: Fun times are bound to end. So keep coming up with new ones.
- At the office or remote?: A bit of both. Mixing comfort with social interaction.
Tell me about yourself briefly
I am 23 years old now, and currently an intern at XWiki, close to finishing my 6 months at the end of August, on the CryptPad project, to be more exact. My hobbies usually include gaming on the side whenever I catch a break. Writing and/or reading for my leisure, in a media literacy sort of way. You could say I enjoy reading or watching text or movies, similar to those YouTube video essays, but I won't post videos anywhere. Just analyze in my head.
Lastly, I enjoy tinkering, though more often than not, breaking my laptop when experimenting with config files on Linux. It's usually just customization, but you could think of it the same way some people enjoy decorating a room. Same principle, plus the need to fix the messy configs afterwards, because I need my laptop to be functional for work and classes.
🌄 The internship experience
How did you find out about XWiki, and what made you choose it for an internship?
I was a fresh graduate, and my interest was always on the cybersecurity/cryptography side of the IT industry. That's generally a smaller pool of availability compared to general software development roles, like in a company that does outsourcing, for example.
I happened to be friends with someone (Daria) and talked about this at the right time when the internship offer was still available. Fast-forward 2 calls, and here I was. As for why I chose it? It suited my thesis like a glove. The internship was focused on post-quantum cryptography, and that's exactly what I had been researching in my third and last year, before applying for Master's studies in the same area.
What was your internship experience like?
It was eye-opening. Due to the nature of cybersecurity/cryptography, there's often quite a big difference between studying or researching it and actively implementing it in a development environment. Before this, my hands-on experience with cryptography was mostly limited to university projects that used standard security measures, like encrypting tokens. The rest was more experimental work, such as implementing attacks to study behavior or testing schemes on my own out of curiosity.
It was a whole different story with CryptPad, a working product. There were many new things to take into consideration, despite working with the same building blocks I was used to. Scale. Computational cost. Efficiency. Since 5 seconds might not be long enough for me to wait for a successful result. But for potentially hundreds of people? It would not cut it.
So I am thankful I got to see and experience the production side of cryptography. I learned a lot of small things to look out for. Most notably, silent failures that tend to move the error you're looking for away from the cryptoscheme itself to some random file you did not even touch before. Frustrating? Sometimes. Worth it? Absolutely.
What did you find to be the most interesting and challenging tasks?
It's more part of the workflow rather than one specific task. CryptPad is an E2E encrypted application, meaning cryptography is involved at pretty much every step of a request being handled. As you'd imagine, suddenly trying to add a whole new layer on top of the preexisting one comes along with unforeseen behaviors and bugs in areas you haven't yet taken a look at.
So I'd say the most challenging part was finding the proper "bite size" to tackle at each step. Big enough to encompass a relevant feature, but not so large it starts causing issues in other corners. That's easier said than done when you're learning the codebase as you add the new stuff in. Had plenty of rabbit whole moments, sometimes even looping around itself. But ending the work week with the feeling that you finally solved or at least accurately pinpointed what the issue was is quite satisfying.
And what about integration into the environment? How easy or difficult was it for you?
The integration was quite easy, I'd say. The first few days were a bit awkward, given that most of my team is remote, meaning the colleagues at the office would often talk about entirely different things concerning their project. But I learned to enjoy that.
You could say it made the communication in the office easier since they would carry out without the burden of common worries about a common thing we were working on. And that translated into topics outside of work, something quite enjoyable during breaks.
As for my remote team, most of the communication would happen between me and Fabrice at first, but the weekly team calls, albeit work-related, eased up the process of casual talk and info sharing on the Matrix channels later. It's 2 different experiences, the remote team vs. the office colleagues, but I enjoyed both sides a lot.
🔎 Open-source impressions
Was open source something you had an interest in before XWiki?
Yes. Perhaps not in the same area of functionality as the ones offered by XWiki, but an interest nonetheless. I use Linux primarily nowadays, so that's the foundation brick, I would say. But even before that, on Windows, I'd tend to scour GitHub first for projects/tools maintained that would cover the needs I had at the moment, before I'd look at the proprietary offers.
So much so, I replaced core utilities for my laptop, meaning fan control, CPU overclocking/undervolting, and power management, with open-source alternatives (e.g., GHelper) over what Asus (my laptop's manufacturer) brought to the table, often bloating or eating up more resources than it was worth it. My only regret is that a lot of open-source things I use require bits of info or skills that I was not at a level comfortable enough to try and help with. Things are a bit different with cryptography, though.
What's the open-source value you identify with most?
The free accessibility. I fully believe any functionality or need should have at least one freely available, secure, and competent alternative. (Since there are free alternatives that do nefarious things under the hood or just not work very well.) It's fine with me if some proprietary, similar app happens to have 1 or 2 features that are better but are closed source.
But I think the idea with open source should be to make the baseline of what's on offer of good quality. If, say, we make a big abstraction and look at proprietary software as one entity, then I think improving open source to a considerable degree will, in turn, kick the proprietary side up its game. It's unlikely everyone will turn to open source for various reasons. But it's worth causing that improvement for the users who won't switch. And if they don't improve, they'll be the ones to lose market share and dissatisfied customers. Not the open-source community.
⌛ The XWiki experience
What would you say is the most valuable lesson you learned during your time at XWiki?
To get a load off sometimes. I think most fresh graduates start off career-wise with this induced formality, either from home or their previous institution. And that can often lead one to be more stubborn than they need to when, say, encountering an issue. Therefore, they bash their head against the proverbial wall just so they don't embarrass themselves or give off the impression they're not as competent as they should be for the job. There are companies out there with this sort of environment. But I feel like that more warm atmosphere here makes it easier to detach and say "OK. I tried everything I could think of. Let's see if my manager has a different idea, one that could help."
At the end of the day, the goal is to solve the problem, not showcase that you did it on your own specifically.
If you could describe XWiki in one word, what would it be?
Patience
Favorite memory at XWiki?
It kind of relates to the most valuable lesson. But it was a similar issue, where I was completely lost, so I asked Fabrice for input to tell me maybe I was doing something wrong with my approach. Only to find out that the issue boiled down to one, admittedly important, file regarding database work. A file that was in a corner, I did not even think to look up, given that the error message was so vague (thanks, build worker), and I was implementing things on the opposite side of the application.
It was a bit like if you had heavy rain somewhere in Britain and that somehow caused the Gobi Desert to flood. It's a favorite memory because I am sure such cases will appear in my future career too. So given that I experienced the frustration already, I will be able to laugh and work much faster next time.
