2024 year-in-review: 'Test it in production' edition

If there’s one thing I’ve learned in 2024, it’s that sometimes you just have to ship it and see what happens. You know, like deploying to production on a Friday, as the tired ol’ cliche goes. For me, this year has been a series of experiments, some successful, some… let’s call them “learning opportunities.” To quote Alchemist from Dota 2: “Failure is just another kind of success. The wrong kind.” (and no, I’m no longer playing, if you’re wondering).

From joining SuperTokens and diving into auth, to co-founding a developer conference, 2024 has been a year of throwing things at the wall and seeing what sticks. And boy, did some things stick.

Here are some highlights:

Co-founded and launched WhatTheStack, because our corner of the world needed a proper dev conference
Kept the local dev community beered and JavaScript’d through 6 BeerJS meetups, and a panel discussion
Shipped 23 articles in total
Spoke at a bunch of conferences and meetups
Dabbled in podcasting and video content (the figurative editing room floor has seen better days tho)

Most importantly, I learned a lot. Mostly by forcing myself to be uncomfortable.

The SuperTokens Saga: A new journey (aka When Auth Met Web Components)

March 2024 marked the beginning of a new chapter in my career. Joining SuperTokens wasn’t just about getting a new job – in a sense, it was a dream come true. Getting to work with an open-source company has been a dream of mine for a long time.

I ended up doing a lot of fun things: building integrations for SuperTokens, recording some videos, doing some talks and writing a lot of articles. Some CLI work on top of that (check out the SuperTokens CLI). And while DevRel is always about wearing many hats, I love the fact that a significant portion of that work was still writing code. Which, brings me to an opinion of mine - even if you’re in DevRel, you still should be a developer first. How else can you advocate for developers, especially when you’re doing it for a developer-focused product?

Opinions aside (and please feel free to tell me why I’m wrong), the real fun started when we faced an interesting puzzle: how do you build authentication UI that works for everyone? And by everyone, I mean that React developer who swears by hooks, that Vue enthusiast who won’t shut up about the Composition API, and yes, even that person still writing jQuery (we see you, and we’re not judging… much).

The solution? Web Components. Not the sexiest choice, I know. It’s like choosing to drive a Toyota Corolla when everyone else is test-driving Teslas. But here’s the thing – sometimes boring is beautiful. This experiment turned into talks at ArmadaJS and React Day Berlin, where I got to convince people that maybe, just maybe, the unsexy solution is exactly what we need. Maybe. If the problem is right.

BTW, we may have a better solution to this problem. Stay tuned 😉

Communities: From Beer.js to WhatTheStack

If what I did at SuperTokens was about seeing what sticks from a bunch of R&D experiments, running community events was about building human connections. You know, the stuff that really matters, if you ask me.

The Monthly-ish Ritual: BeerJS

No, I'm not taking your React Components away, don't worry.

Six BeerJS meetups might not sound like much, but that’s six evenings of bringing developers together, sharing knowledge, and yes, occasionally debugging code over a pint. Live, on stage, of course. Besides being a household name by now (not my words, I’m quoting someone here), these meetups became our testing ground for bigger ideas - a way to keep our finger on the community’s pulse and understand what our developer community really cares about.

More on our BeerJS journey here

Which, leads me to my next point.

The Big One: WhatTheStack

They'd let anyone on stage these days...

Remember that “test in production” theme? Well, nothing says “YOLO deployment” quite like starting a developer conference from scratch. WhatTheStack wasn’t just another tech conference - it was our attempt to put our corner of the world on the tech community map.

The conference was built on a simple premise: Macedonia deserved a world-class dev event. No more “it’s good enough for a local conference” - we wanted something that could stand shoulder to shoulder with the events we’d been attending abroad. And, we wanted to build something community-first, affordable and ultimately something that you’d really be looking forward to come to, year-over-year.

Some highlights:

Multiple tracks covering anything and everything webdev. Even some tangentially related stuff - we’re fairly flexible like that
International speakers sharing the stage with local talent
A proper venue (because our local community hotspot fits about a 100 max)
Real sponsors that actually gave us money (turns out, people sometimes will support your crazy ideas)

The best part? It worked. Despite the sleepless nights, endless meetings, and that one moment when we thought everything would fall apart (there’s always that moment), we pulled it off. The community showed up, the speakers delivered, and somehow, we managed to create something bigger and better than we imagined.

Details here

Content Creation: Writing - good; Video - mid at best; Speaking - decent, gotta do more

2024, for me, turned into a year of writing, with 23 articles split between my personal blog and the SuperTokens blog (6 on my personal blog, 17 on the SuperTokens blog). To some, that may be a little. To me, it’s a record. However, my video efforts were quite mediocre at best. I did okay with speaking though.

Writing

The Personal Corner (darko.io)

I try to write as much as possible. I should probably do it more - 6 is not a lot. Some highlights with author’s notes:

BuT, aUtH iS HaRd - I love this piece. It was inspired by a discussion I had at a conference, in which someone vehemently argued that auth is hard, and no-one ever should build one, period. Just use a service. Now, that person may or may not have been involved with a company that builds auth, but that’s not the point. Funny enough, it’s not that a completely disagree - I’m very much pro re-using solutions to obviously solved problems. But I hatehatehate gatekeeping. And who knows, you may end up inventing a better wheel anyway 🤷‍♂️
The Joy of Astro - I love Astro. Not much more to say. Especially when it makes my life easier in building a SuperTokens integration.
Web Components: The Unsexy Solution to Cross-Framework UI - conference talk turned blog post. A more detailed version of the topic.
So You Want to Start a Developer Conference? - sharing the raw, unfiltered story of WhatTheStack. And if you’re wondering whether to start one, yes, consider this your sign to do it.

The Professional Stage (SuperTokens Blog)

At SuperTokens, I got to write a lot of technical stuff. Sometimes SEO driven. I’m not a huge fan of those, but they aren’t a ton of work for me - they are kind of formulaic anyway. I managed to write 17 articles that ranged from technical tutorials to comparisons of different auth solutions. But here’s the thing - any kind of writing practice is good practice. Do I love all of those pieces? Absolutely not. But that doesn’t really matter all that much. Some highlights:

Experiments in Audio and Video

While not as consistent as I’d hoped (hello, 2025 resolutions!), I dipped my toes into podcasting with CodePub. To be fair, I kinda started doing that a while before 2024. I consider this one of those “learning opportunities” mentioned earlier.

The key learning from all this content creation? Consistency beats perfection. I need to learn consistency. I also want to start streaming my 1001 weird project ideas. One way to really commit to shipping one of them is being publicly accountable for it. So I hear, at least.

Not much else to add really - check it out here: CodePub

Speaking

Compared to how much I couldn’t be bothered to even apply on CfPs in ‘23, ‘24 has been a stellar success. My talks, panels, MCing etc. were pretty fun. Things are already looking better for ‘25 on the speaking front too - so expect to see me more in the wild. Some highlights from ‘24:

I swear, if I use "like" as a filler word one more time...

ArmadaJS - Web Components, mentioned above. And probably my most enjoyable conference experience in 2024.
React Day Berlin - Same web components thing, shorter, also mentioned above. Ended up doing it online (not as fun, but it was what it was). (Recording here)
WhatTheStack - I got to give a keynote. Well, frankly, I got my hand twisted into doing it, since I’m really not into speaking at conferences I organize. But it was fun (recording here).
WeAreDevelopers World Congress - I MCed stage 2. For two days in a row, even though I was warned against it. Even though MC-ing is filler content, you get to spend more time on stage than any of the speakers. Plus, you’re also stuck on your stage for the entire day. But, it was fairly educational. My personal highlight from that experience: A good stage manager and crew are worth their weight in gold. Even though the experience was tiring, it was easy compared to every other MCing experience I’ve had, all thanks to the crew and stage manager.

The most fun part of any conference - beers afterwards.

Lessons from 2024

Looking back at this year of “testing in production,” a few key lessons stand out:

What Worked

Sometimes the unsexy solutions are the right ones (looking at you, Web Components)
Taking chances on big ideas (like starting a conference) can lead to amazing outcomes
When in doubt, ship it. And no, this doesn’t mean you’re supposed to ship dysfunctional crap. It’s a balancing act.

What Needs Fixing (2025 TODOs)

Actually commit to a regular cadence with streaming/video content.
Write more on my personal blog.
Actually commit to shipping one of my 1001 weird project ideas. I’ve been toying with this D&D assistant LLM that’s helping you with making NPCs feel more alive… that might be a good one.
Speak more (and actually put in the effort to apply to more CfPs).

The Curtain Call

If 2024 was about testing in production, it was also about proving that the best way to build something meaningful is to just start building it. Doing the thing is the only way to, well, do the thing. Whether it’s challenging the “auth is hard” dogma, bringing international speakers to our corner of the world with WhatTheStack, or maintaining the rhythm of BeerJS meetups – everything started with a simple “someone’s gotta do it”. And realistically, you have two options here: wait for someone else to solve it for you, or do it yourself. Only one of those is deterministic.

Here’s to 2025 – may it be filled with more experiments and continued growth. And remember, if something breaks in prod, literally or figuratively, that’s what hotfixes are for, right?