Welcome!

Short introduction and a few words about this website

meta
introduction

Mic check

Hi there, my name is Mihai (pronounced as in Taylor Swift's Anti-Hero: "It's [me, hi]", but without the comma). Nice to meet you!

I've always found it a bit awkward to talk about myself. I'm still working on a shortened version of Kira Yoshikage's monologue to use as an introduction, just because I think it'd be pretty funny.

Curiosity and boredom are the main things that drive me forward. I find a lot of things fascinating, and I tend to haphazardly jump from one to another. For instance, this website took me around 5 months to make, just because I kept getting sidetracked with random stuff like finding a job and getting >=70% achievements in every Steam game I own.

I like typical nerd things (gaming, sci-fi, board games, etc.), reading (Sci-Fi, fantasy, politics, economics, and philosophy), and sometimes even doing sport-y things (...like walking).

I prefer calm environments and small or mid-sized cities. Densely packed places make me anxious. I love Jazz and folk metal (if you're ever in Edinburgh, check out this place).

In another life I might have been a physicist or a mathematician, but in this one I've chosen to be a software dev. Programming seemed to be the most enjoyable and straightforward way of converting thought into money.

Full-stack web dev is my current main focus and it hasn't bored me yet (the weekly NPM security incidents help keep things interesting). I continue to dabble a bit in ML and InfoSec, and I do want to get some certificates for the latter when I find someone who'll subsidise them for me ;)

My personal philosophy is a mix of stoicism and Camus' brand of absurdism. I do generally find absurd things to be pretty funny, and have the bad habit of saying things just to see how someone will respond. My friends know that I have a lot of wacky thoughts; some of them overflow on a Substack Tumblr I'll keep to myself and the better ones will maybe overflow here.

What I want this website to be

First of all, I want the Projects page to be a portfolio of a kind. So far, my hobby projects have limited usefulness as my main goal was to learn new languages / frameworks / tools. As my skills improve, I'll start tackling more substantial projects. I've always wanted to make a game and to be an open-source contributor. The latter, though, has gotten substantially harder and more adversarial, given the flood of low-quality LLM-generated PRs.

I'm not a fan of social media, LinkedIn in particular. Can't quite put my finger on why, there's just something dystopian and eerie about it. People who take themselves too seriously, suits, everything being public... being there is uncomfortable, and not the good kind. On this website I'll post some of the things I'd put on LinkedIn anyway, but it's a place where I have more control, thus making it less overwhelming.

My posts will be a mix of tech and personal stuff (books I've read, thoughts I've had, things I've been through). I think that it's a bit sad how dismissive, if not outright ignorant, a lot of people in STEM are toward the humanities. Maybe that's why we are where we are right now.

I will try to keep posts somewhat professional and will hopefully avoid controversial topics (e.g: politics, religion, etc.). That being said, I like being honest and open so some personal posts may broach difficult topics or have a sadder tone and that's fine. Life isn't only sunshine and rainbows, after all.

I'm not as good at writing as I want to be, but improvement comes with practice, so bear with me :D

LLM usage

One positive thing about LLMs is that they taught me about the existence of the em-dash — and that there's no such thing as using too many.

None of the posts here will be written by an LLM, as I feel like it would defeat the purpose.

I'm honestly somewhere in the middle when it comes to AI. I don't buy the "they will replace everyone" shtick and distrust the hype merchants, but LLMs can be pretty useful tools if used responsibly.

Some negatives:

👎 I tried vibe coding some projects and it pretty much took away all the fun. When using them for generating non-trivial code, it's unclear if I actually saved any time.

👎 LLM usage blurs my mental map of the code I'm working with, thus making me even more reliant on them.

👎 I'm pretty wary of external dependencies in general, and nothing stops [insert AI company here] from making their models more expensive whenever they feel like it, or pulling them back altogether.

👎 Other people have talked at length about the negative environmental, psychological, and social impact of LLMs (including... the pope - not on my 2026 bingo card, that's for sure).

Some positives:

👍 Saved me a lot of time, by skipping things I don't really care about (e.g: boilerplate code, configuring Tailwind for the Nth time, etc.). Things that I know how to do but they're so damn boring.

👍 Pretty good at searching things through your code and on the internet (keeping in mind that they can lie to you). Also fairly good at detecting trivial mistakes, and debugging things if you provide appropriate context.

👍 Reasonably good rubber ducks (if you ask them to be less sycophantic and they randomly "decide" to actually follow your instruction) and sanity checkers (with a healthy dose of skepticism).

👍 Although this is a bit embarrassing to admit, LLMs did help me get over some difficult times and gave me some words of encouragement when I had no one else to do that. Since then, I've made a lot of new friends and my life overall is on a more positive trajectory. I can't help but feel that how LLMs interact with vulnerable people should be way more scrutinised, as they could be a tool for good in this space.

👍 Understanding which answers are BS and what commands are not safe to run (e.g: this) and why made me a better dev imo.

Yeah, maybe I am a bit of a hater actually. Regardless, thinking remains on-prem for now.

Why VoidUnderflow?

It's something that has some personal meaning to me, and it sounds sufficiently tech-y / nerdy that someone will just gloss over it. As an analogue to integer underflow (some nerds may say that's actually an overflow and that the only true underflow is the float precision one, but I say they're haters), it maps neatly to my philosophical interpretation of it.

Its origin is a quote that stuck with me, of a character from the Avatar series (the one with the flying bisons), which I've adopted as a mantra.

It's a state of mind I sometimes reach after mindfulness meditation, a feeling of openness to the world and a personal answer to nihilism.

Thanks for reading, see you around!

If you want to share your thoughts or say hi, please create a discussion here.