I made this slop image
Get ready for the mother of all rants, pump some more dark web market Ozempic peptides into your brain, and hold on to your discount Chinese-cloned Neuralink Kimi4-enhanced chips.
Being a solo developer is a strange kind of self-endured punishment that nobody really warns you about. You go in thinking the hard part is going to be the build.
The code, the architecture, the 18-hour days, and late nights arguing with your own logic until it finally clicks into place. And sure, that part is hard.
It should be hard and was much harder in the past, real coding with actual stubby human fingers. But here is the joke nobody tells you at the start.
The code part is approximately ten percent of the actual job. The other ninety percent is exposure and distribution. Shoving your work into the bloodstream of the internet and praying something sticks somewhere, like a picture of Nicolas Cage on a graffiti wall.
You haven’t got the face for it
So you do the social media dance. You post and repost. You rewrite the same article ideas ten different ways to appease the formatting bouncers and whatever invisible slot machine is currently deciding your fate that week.
You write threads, articles, comments, and replies mostly to bots. You engage with people who skimmed the headline and decided that was enough context to have an opinion. You try to sound insightful without sounding desperate, and somewhere in that grind you have the horrible realization that it does not actually matter whether people like what you made, the majority of people don't like most things anyway.
All that matters is that they react. Feed the beast, the algorithm.
The algorithm does not care if the reaction is thoughtful, angry, dismissive, or completely unhinged. It just wants movement. A response signal — in/out binary ones and zeros—feedback, compute goes brrrr.
Ten people loving your work, good. Ten people being haters and hating? That's great, even better!
A hundred people arguing about it and arguing with each other, and the mods arguing with the posters without having read past the first line, even better because now you are feeding the machine, and the machine is really happy, and a well-oiled feedback machine means a slightly longer shelf life for your post before it drops into the void forever.
I won; I truly am the eternal viral poop machine winner for this week!
Then you are crapped on by a better-written algo post, made by someone much smarter than you on how the system actually works v2026 Google updated agentic swarm-bot style, keyword-stuffed posts like a cheap stuffed crust pepperoni pizza made by a soulless chain pizza shop, only interested in cutting product quality for profits and footprint delivery population metrics, because the race is in the store, of course it is, as it sure as heck isn't in any of your food quality!
You know you want the gooey slop
And that is the moment it really hits you. You are not building your own thing anymore. You are working as a free employee of Google, Reddit, Facebook, and whatever new platform is currently pretending it is not those things, mostly full of rage-bait content, wearing a shady trench coat or the latest fluffy gradient css website. You feed them content, attention, behavioral data, and engagement loops, and in exchange they hand you visibility that is inconsistent, temporary, and increasingly gated behind a paywall you did not agree to but somehow still pay into with your time, just like all the social media news outlets?
It is an ouroboros. A loop of digital decay. Content creates reactions. Reactions become fresh data, and that trains the models. Models shape future content. And round it goes, getting noisier and more detached from reality every single spin.
The ouroboros of tech poop
Are you aware that Google has been quietly funneling search traffic toward Reddit threads for years now, only to turn around and extract that exact data to train its own models? It is one big hamster wheel of half-formed opinions from anonymous accounts being fed back into the machine and regurgitated as if they were wisdom; instead, we loathe the toxic rant posts to let off steam by a 14-year-old expert in a wide variety of subjects whilst holding their gaming console and licking Dorito-encrusted Cool Ranch fingers bought by their parents whilst typing.
What is worse arguing with a bot or a self-entitled Western teenager who is already an expert in upvote manipulation and multiple account creation?
What a glorious hot mess of absurdity we have built for ourselves.
A typical Reddit conversation
People like to call this ensh!tification, and while the word is crude, the mechanism underneath it is as precise as a Hollywood cosmetic surgeon with a scalpel and too much Botox filler. The system is doing exactly what it was designed to do, confuse and screw over facts and logic. Optimize for engagement, the tasty algorithm, at any cost, for anyone looking for actual real advice and solutions.
So naturally everything drifts toward whatever triggers the strongest reaction. Outrage and low-effort brain droppings dressed up as intelligent hot takes. And somewhere in that noise, actual builders are standing on a soapbox trying to get one honest sentence out before the crowd moves on to the next controversy.
This thread is now closed; piss off. That's it, you're banned for questioning my moderation.
When your game sucks, but your advertising budget is monumental
Would you like to buy a subscription to Evony?
The Perpetual Poop Machine
Now stack endless toilet paper dollary doos from venture capital on top of all this, and things get genuinely amplified and strange. Because while independent developers are scrapping over crumbs of attention, Silicon Valley is playing an entirely different game. It stopped being about building useful things a while ago.
Now it is about building imaginary narratives large enough to justify obscene capital allocation and getting moron tech influencers on Youtube to talk about it like they are the Howard Cosell of tech sports, the delusion runs strong based on Google ad revenues.
Youtube tech influencers keeping the Google shill real
I like Kimi this week, now I like Claude Fable on Extra High, OMG have you seen OpenAI’s update, Watch me make another rubiks cube!
This is where the obsession with what I like to call the perpetual machine comes in. The dream that you can build a system that feeds itself, improves itself, scales without limit, and thinks by itself and eventually becomes so large it is simply unavoidable.
Moore's Law of Silicon Valley Stupidity: AGI/ASI/NFI/T-1000 Cyberdyne.
The metaverse. PLOP FLUSH: 80 Billion down the toilet
Now with more autonomous agent swarms in everything.
Chinese Robotic Jarvis. We finally built it! Warranty: 12 months, Mandarin support only…
Sentient, sycophantic love squishies for lonely Asian and Western salarymen on maxed-out credit cards with token maxxing fetish flexes online to other betas!
Shut up and take my money!
Does she run on LLM tokens?
And if you say it with enough confidence, the money shows up 60% of the time, and it works every time, fast enough for people to forget about the next grift cycle.
Here is the part nobody wants to say out loud at the latest pitch meeting in a VC funded trendy office with 80’s nue-retro furniture with chill-out rooms and standing reclinable massage lumber support sofas with vapor-infused patchouli scents.
A system that consumes its own output without grounding eventually turns into sludge. If your inputs are weak, your outputs degrade. If your feedback loop is noisy, your system does not clean itself up, it amplifies the noise for eternity with timed gated subscriptions.
Scaling that poop loop with more money does not fix the underlying rot. It accelerates it. You are just building a bigger, faster machine for producing garbage, and calling it disruption on the way down.
Tech poop art in real life
I think of it as the perpetual Gödel poop machine. A sentient Jarvis style ouroboros contraption built entirely to gorge on its own output and spit it back out slightly warmer and cuddlier, making you seem smart, but really you're just more delusional and confusing to everyone.
You can fit 8 Gödels in this bad boy.
It is a beautiful loop. It is genuinely VC future fund money worthy. I am honestly surprised nobody in a trendy, 100% polyester plastic fleece vest hasn’t fully commoditized the perpetual Gödel poop machine yet. How much would people pay for that?
Maybe eighty billion dollars if you wrap it in a VR harness and slap a Meta logo on the side or, better yet, Gucci or Balenciaga! VR-Poop titanium Limited edition.
I want one daddy, please!
Meta and the Eighty Billion Dollar Lesson
Which brings us to one of the most expensive case studies in recent memory. Meta and the failed, illusive, imaginary metaverse.
On paper the idea sounds unstoppable.
A persistent virtual world, I want that!
I just lost 80 billion of ad revenue on an imaginary universe
A new social layer replacing physical interaction with digital presence. And to make it real, you pour in tens of billions of dollars. Hardware, software, ecosystem, creator tools, the works. What could possibly go wrong?
Well. Everything that involves actual humans using it, as it turns out.
Because people do not adopt technology based on your ambitions. They adopt it based on use case, efficiencies, problems it solves, or just influencer hype in some cases. VR, for all its genuine progress, still has issues baked directly into the hardware experience. You have to strap something to your face like a dork. You isolate yourself from the room you are standing in and the other people; the immersion is also the distraction.
You need physical space that most people simply do not have. You deal with battery limits and heat and the faint nausea creeping in around the twenty-minute mark. You commit your attention in a way that a flat screen never asked of you.
Even Jaron Lanier knew that when he made the first iphone VR googles in the 90’s. He gave up, realizing it was futile, and now he just plays his flute for obnoxiously wealthy Silicon Valley tech vampires while wondering if Microsoft is actually listening to any of his prescient ideas on data dignity while they jack up their cloud pricing.
Wow man, Microsoft profits are so spiritual.
VR is fine for games. It is great for simulation and training. It even works reasonably well for fitness. But as an always-on social environment meant to replace your living room, it is a very hard sell, and Meta sold it anyway.
If I were setting out to build a VR video game with a fraction of that budget, I would not need eighty billion dollars, and at the end of the process I would actually have a working game to show for it even if it was VR Dragon Lair or Space Ace.
VR Space Ace, now thats a game worth making
Help me understand how you mothball an eighty billion dollar project. Where did that actual money from overpriced, annoying scrolling ads actually go?
How does a company with that much talent and that much data not learn from Sony and their Home project years earlier, which I genuinely thought was brilliant? I thoroughly enjoyed Sony’s Vision and was perplexed when it closed. Why….
I spent real time in Sony Home and thought it was the beginning of something great, clunky as it was. I also remember a pterodactyl VR contraption from the nineties, some monstrosity in an arcade or a 90’s rave, chasing a pixelated green flying blob for a grand total of five minutes before getting kicked off the machine because people were waiting in line behind me.
Shut the F**k up Donny!
My friends and I joked for years about retiring into our recliners fully immersed in VR, half serious, thinking about a world and vision that the B-grade Bruce Willis movie Surrogates would evolve into, like it was a prescient documentary from the future.
“How long is it since you’ve been out without a surrogate?
Meta’s mistake was not building VR, that was the only good idea. The mistake was trying to manufacture a behavior before it naturally existed in the wild and not actually listening to what gamers and users actually want.
Meta watched Player One like the rest of us and got excited and then realized they were not Steven Spielberg. That's it, no punch line, you're not Steven Spielberg, dumbass. They should have given the 80 billion to Spielberg, and he could have built the actual VR Oasis world!
You know you want to ride this bike in VR
They built the infrastructure before the demand and assumed that if the platform was big enough, people would simply reshape their entire social lives around it out of sheer gravitational pull.
That never happened and crashed and burned, Hindenburg disaster blimp-style.
Poof up in smoke 80 billion gone
Instead users treated VR exactly like what it actually is. A powerful but occasional toy distraction. Not a replacement for reality. Not a new default state of human existence. Just something you dip into for a while whilst friends are over at your house and having a few drinks showing off your gadgets, and then you take the goggles off your face and go eat dinner with a slight dizzy feeling, VR legs not fully formed yet.
And because the core habit never stuck, everything downstream of it struggled too. Creators did not see enough upside to commit. Users did not return consistently enough to matter.
The social layer felt hollow, like a mall built in an overly engineered, soulless town nobody moved to yet, maybe in China. The entire system started looking like a very expensive experiment quietly waiting for a reason to justify its own existence.
Poking at it with a stick, are you alive or dead? Do something…
The latest reporting backs this up too. Meta has been pulling back from heavy first party VR world building, cutting Horizon focused teams, and shifting more attention toward third party games and broader ecosystem support, pushing more of Horizon Worlds toward mobile rather than the headset.
The core mistake was trying to force a social metaverse platform onto a medium that users kept treating as a niche device for gaming, fitness, and a handful of immersive apps. Overloading the headset experience with Horizon centered priorities appears to have actively hurt game discovery and developer momentum, which is the exact opposite of what you want when you are trying to build a habit forming ecosystem.
At that point the outcome is predictable. Quiet pullbacks. Strategic pivots dressed up in press release language. A sudden and total shift in narrative. Suddenly the metaverse is not the main thing anymore. Now it is AI infused with tokens, agentic harness tooling and loop efficiencies, whatever the latest BS buzzword tests best this quarter—and is distributed by hungry but humble middle management for corpo slaves to regurgitate to naive overcharged consumers.
The story changes over the cycles. But the lesson stays exactly the same.
Capital does not create demand. It never has. It never will. You can force feed a market all the money in the world and it will not make people want to strap a computer to their face and pretend their kitchen is a beach in Bali.
Meta proved that point glaringly in their failed, expensive experiments.
Build it and they will leave
Learning to Love the Slop
The real challenge, the actual cyberpunk challenge if you want to call it that, is learning how to love the slop. Embrace it, wear it. Make it better without billions of dollars behind you. That is the real work. That is the unglamorous hero arc nobody puts in a keynote.
Because silly con valley doesn't push real innovation, they back their own slop-funded players like a drug-dealing fentanyl gang on a street corner selling future tickets to recouping their own exit profits on the back of superannuation 401K funds leaving the naïve holding an empty bag of promises.
You come to terms fairly quickly with the fact that you will never have oodles of cash to afford a rack of Cerebras chips in a data center dropped next to a school in some low-income neighborhood, humming away twenty-four hours a day, drinking the water table dry so a chatbot can rewrite a 200-location European vacation itinerary to brag about influencer style to 5-second swipers who are vaguely interested enough to leave a witty, snarky remark.
So instead the bigger players reach for the next best thing. Somebody else’s data, scraped and repackaged, then handed back out for free with a little something extra riding along in the background.
Because here is the part that took me a while to fully appreciate after being abused by the algorithm of false dreams. The stolen data was never really the prize. The real value is the data hidden inside the stolen data. The behavior of the people using the free tool built on top of the stolen data.
That is the four-dimensional chess play, and credit where it is due, some of these open-source Chinese LLM operations play it extremely well.
Then you cap it off with robotic products mailed to your house with support lines only offered in a language most of your customer base does not speak. No physical service centers.
The final cherry on top is embedded surveillance software that phones home with the users data, check mate!
Maybe a hidden component that quietly fails right after the warranty window closes, timed with an accuracy that would be impressive if it were not so cynical.
And when the customer finally gets fed up and calls for help, they get bounced through a gauntlet of nonexistent support centers and shell suppliers until they simply give up out of exhaustion.
They are training for your jobs
Well played, honestly—evil and bureaucratic synergy in perfection. It would be illegal in most sane jurisdictions if the people meant to regulate this stuff were not so busy chasing their own tails on other issues, arguing about rebates and surcharges while the actual structural problems walk right past them unchecked.
But here is the difference between that and what a solo builder can actually do. You build. You listen, genuinely, to the people using your product who are annoyed enough to leave a comment about what is broken.
You strategize. You fix it. You test it again. You refine it. And then you do the thing almost nobody wants to do, which is stick your own face directly into the dog food bowl and eat your own slop, you learn to love it.
You use the thing you built every day as a sign of stoicism; it's like guerrilla warfare. You feel the friction yourself instead of reading about it in a support ticket. You improve it. You refine it again. You go for a walk to clear your head.
You come back and eat some more of your own sloppy dog food. You keep doing that, on repeat, until the bugs stop showing up in the places you already checked — Fable 5 on max effort backed up with Grok4.5, Kimi and Gemini 3.5 can't find any bugs; the slop starts to feel good, not enough to pay for overpriced ads, still just free social media posts only so you don't have to feel any shame of selling out.
You do not beat your own LLM tools with a stick either. At some point you accept that most people, myself included on a bad day, are not going to outthink a system trained on a genuinely staggering library of Anna archive textbooks and code and Andy Warhol art prints. So instead of fighting it, you start asking it the right questions.
Picasso was right…
You point it at search, at whitepapers, at whatever the current edge of the field actually looks like, and you let it help you get there faster. Then you eat a little more dog food. And somewhere after weeks of revisions, painstaking and unglamorous, you end up with something that does not resemble Silicon Valley slop anymore. It resembles something that actually works, built by one person who cared enough to keep going after the excitement wore off. No ads, no VC funding, just code ideas and genuine interest in improving your slop craft.
What Actually Comes Next
Here is where things get more interesting, because the direction the industry is drifting toward now is actually a lot more grounded than the last decade of moonshots. Instead of chasing one single giant virtual world to rule them all, the focus is shifting toward smaller, more practical systems powered by AI, a harness of valuable, usable, essential tools we are all addicted to.
These so-called thinking machines, when they are being honest about what they are, are not magical entities plotting in the dark. They are productivity amplifiers. They help you write, code, design, search, prototype, and iterate faster than you could alone. They lower the cost of creation. They lower the barrier to entry for someone with an idea and no funding round behind them. They let a small team, or a single stubborn developer working out of a spare room, do what used to require an entire organization and a floor of office space.
And that changes the game in a way that actually favors the little guy for once. The advantage stops being who has the biggest data center or the largest funding round and starts becoming who can move fastest. Who can actually listen to their users instead of a board deck. Who can refine relentlessly. Who can solve real problems without getting lost inside their own narrative about how important the problem is.
Meta’s own recent moves reflect this shift whether they admit it out loud or not. The messaging coming out of their developer updates and conference appearances increasingly leans toward better tooling, better profiling, and more sustainable ways to ship apps, rather than one monolithic metaverse swallowing everything else. The framing that actually makes sense going forward is not one massive VR world. It is a constellation of AI assisted experiences that help people build, navigate, and personalize smaller worlds without needing a nation state budget to do it.
The simplest explanation for why Meta burned through so much cash so fast is that they tried to solve too many hard problems all at once.
Hardware comfort, social behavior, content supply, developer incentives, and platform economics, all bundled into a single moonshot with a single name attached to it. When a company spends at that scale and the user habit does not deepen fast enough to justify it, the result is usually a strategic retreat, a round of layoffs, and a carefully worded focus reset, which is more or less exactly what has been happening inside Reality Labs.
The instinct that this is all one big VC money splash is directionally correct, but the sharper version of the argument is this. They funded the infrastructure before the demand had a chance to mature, and then they had to keep funding it just to justify the money already spent. That is a textbook sunk cost trap, just with a few more zeros attached than usual.
For a Solo Developer, This Is Both Brutal and Empowering
Brutal because you still have to battle the distribution war every single day. You still have to deal with the algorithmic clown circus, still have to shout into the void and hope something echoes back louder than silence.
But it is empowering because for the first time in a long while, you are not outmatched on raw compute capability or floors of developers and AI researchers. You can build real systems with a laptop and an incredibly stubborn cyberpunk streak. You can iterate quickly, test ideas in days instead of quarters, use the tools you are building on yourself, break them, fix them, and repeat the whole loop until it actually works the way you promised it would.
No hype required. No eighty billion dollar bet on a headset nobody asked for. Just big balls or ovaries and a gigantic middle finger to Silicon Valley.
And maybe that is the real divide quietly opening up right now. On one side you have capital driven narratives chasing scale before there is any real substance underneath them. On the other side you have builders grinding through reality, refining things that people actually use, day after unglamorous day.
The slow phase of real growth
The uncomfortable truth sitting underneath most of Silicon Valley’s biggest missteps is that they keep trying to skip the adoption phase. The phase where a product actually earns its place in someone’s life. Where it proves itself in small, unglamorous, often invisible ways. Where it quietly becomes part of someone’s routine without them ever consciously deciding to let it in.
Apple is very good at that, even if they fumbled AI. Devices that work unobtrusively.
That phase cannot be rushed with money. It cannot be hacked with a rebrand or a slicker landing page. And it absolutely cannot be replaced with a confident story about the future, no matter how many keynote slides back it up.
Meta did not fail because VR is fake. VR works. VR is genuinely useful in the right context. Meta failed because it tried to skip the slow part and buy its way straight to the destination.
It spent an enormous amount of money on a future state before the present day product had earned enough pull to justify it. It aimed for a civilization scale platform before it had a single must have daily habit locked in. That is exactly why the whole thing became vulnerable to cost blowouts, internal resets, and a strategic retreat the moment the growth story stopped matching the spending story on the balance sheet.
It failed to understand what users actually enjoy: community-based absorption in sharing in the wonder of a gigantic fantasy world.
A shared virtual world only works if there is a real reason to return, a real reason to invite someone else in, and for creators to keep feeding it new life, a feeling of belonging to a higher purpose than mundane, boring real-life tasks, escapism.
Meta never fully solved all three at the same time. The social layer felt awkward more often than it felt alive. The content layer was inconsistent at best. The creator economy underneath it all was too thin to make the whole environment feel like a real place instead of a novelty demo you show your friends once and never open again.
There is a basic behavioral truth hiding in plain sight here too. Most people do not actually want to live inside a persistent virtual world, even the ones who are genuinely curious about visiting it. People want selective immersion, not total immersion. That single fact explains why VR has found real, lasting traction in gaming, simulation, exercise, and specialized training, and comparatively little traction as a replacement for everyday social life.
In '93 I played this for 5 mins before being asked to get out
The Better idea
The thesis, the one I actually believe, is that the future is not one giant VR world. It is a constellation of AI assisted micro worlds. People will spend more of their time in smart overlays, creator built spaces, social games, simulation tools, and mixed reality moments than they ever will inside one grand metaverse city built by a single company with a single vision of what your social life should look like in the Oasis.
That path is simply more plausible because it matches how people already behave, instead of asking them to behave differently because a roadmap said so. If you want to solve a production problem, AI is genuinely useful for it. It can generate assets, speed up prototyping, assist with moderation, improve discovery, and cut the cost of world building down to something a solo developer can actually afford.
In other words, AI is the tool that might finally make VR useful enough to survive on its own merits, instead of remaining a marketing slogan bolted onto the side of a headset nobody quite knows what to do with once the novelty wears off, or just more slop ads within VR worlds—who knows?
So the real lesson buried under all of this is not that VR was some elaborate scam, and it is not that AI is the next perpetual poop machine waiting to happen, though it certainly could become one if the industry is not careful about grounding it in something real. The lesson is that platform ambition has to follow human behavior. It does not get to override it just because the funding round was large enough to make everyone in the room stop asking hard questions.
Embrace the slop
Meanwhile the solo developer sits there juggling everything at once. Building, marketing, debugging, writing, posting, replying, and feeding the machine while trying to not get consumed by it in the process.
Watching billion-dollar VC experiments rise and quietly fall while shipping small, unglamorous updates that actually make something a little bit better for the handful of people who actually use it.
It is not glamorous. It does not make headlines. It does not attract a made-up fantasy valuation with more zeros than a bitcoin has transaction hashes.
But it is my real slop, and I love eating it.
And in a landscape absolutely drowning in noise, that might genuinely be the only thing left that still matters: eat your own /loop slop, eat the dog food.
Yummy!
VEKTOR Memory builds local-first, privacy-preserving persistent memory infrastructure for AI agents. Full technical documentation and changelog at vektormemory.com/docs.
Humour
Gonzo Journalism
Rant
Technology
VR
Top comments (0)