Product upvotes vs the next 3

Waiting for data. Loading

Product comments vs the next 3

Waiting for data. Loading

Product upvote speed vs the next 3

Waiting for data. Loading

Product upvotes and comments

Waiting for data. Loading

Product vs the next 3

Loading

vibecoder.date

Find who you vibe with, git commit to love

A dating app for vibecoders. Swipe and chat right in VS Code, Cursor, or Windsurf. No tab switching. No context switching. Just vibes. git commit to love.

Top comment

I'm sitting here, hunched over an overheating laptop with bad posture. Engineering is exhausting, truly—especially when you've grown to care about it. I don't think I ever experienced vibe coding as magic. The capabilities astonished me, sure. I've been surprised and delighted again and again. But I never felt I was vibing. I was planning, thinking about specifications, constraints, just one layer of abstraction above code. Maybe two. It dawned on me that the reason vibecoding took off is because for once, you can state a desired outcome—like "make me an app that organizes my day"—and just through conversation, it works. Sure, there are limitations. Can't exactly ask Lovable.dev to build AGI. I tried it and all I got was a stupid signup modal. I still wanna try again but I might get banned. As someone who has been in many parts of tech—support, QA, backend, and a lot in between—it was never magic to me. Or rather, the magic was seeing how it worked, not that it worked. It started as a joke among friends. "Oh what if there was vibecoder Tinder?" Well, why not. I bought the domain on a late night whim. Everyone thought it was funny. It sat for a while. Truth is, I've been building things for a long time. It gets isolating. Lonely. You can't open normal conversation with "Oh by the way, yesterday I was thinking about making a Python library to automate unittest creation by navigating execution paths." I mean, you can, but even to the technical people it just... feels like noise. Engineering is being jaded as an applied science. But I kept thinking about it. The meme was funny. It kept me going. What if you could chat from Cursor? Or just connect? And yeah, the dating aspect is gonna be the focus, just because it's fun. YOLO and what not. I started building. Not with any sort of engineering expertise or direction in the aesthetic beyond "I'm vibecoding." Of course some good practices slipped in—double checking, reviewing work, considering backend providers—but even then, the time it took to act was much less. And I'm seeing this shift. Not in enterprise, at least not yet, but in the Twitter space. Where people are making projects that we once deemed too time consuming to matter. Even this blog is vibe written. I usually keep all my writing that isn't purely functional away from AI, but... this is the vibe people talk about. Messy input in, structured output out. Complex problem in, software solution out, iterated over, refined. But the choices are now at a level of functionality and usability, not architecture. I didn't choose to use React for this, not a priori—I asked and it was something coding models do well. I'm using Supabase because their MCP is pretty good and I don't have to worry about anything being done by hand. I'm deliberately letting go of the usual framing around software. No thinking like an engineer or a product owner or any formal role in tech. It's now: I want the app to do X, how can it be done, okay, do it in Y way please, okay now check that things are done securely. Notice: at no point am I even thinking of design patterns, or frameworks, or tech debt. The principles become very user focused—usability, experience, how does it feel to interact with. Vibecoding kinda eliminates the need to plan ahead. Not as in carelessly charging in, but as in I no longer need to make sure I can expand the codebase and make it scalable from day one. Now we can majorly rebuild what we need when we need it. We don't need to plan for coding velocity or onboardings. Not for an app like this anyway. Planning isn't about being able to scale from the get go. It's about being able to rebuild enough when you need to scale. When this app gets enough users to start creating slowdowns, I know I can make it scale—not because I'm a good engineer, but because I'll have enough data to vibe architect towards the needed throughput. Is this the way all software will work? Fuck no. Some software will always need to be deterministic and checked manually and with machine help. Some software needs to be reliable. But one 20 minute outage here isn't gonna kill anyone. It's not gonna crash half the internet or cause property damage. It's just gonna get patched. I'm usually highly deliberate with my writing, but this time, it was all vibes. No structure, no nothing. I expressed what I wanted to express and I might not even look at the formatted output. Before YOLO there was carpe diem, and before that, fuck if I know. But that doesn't matter now. This life is more than being procedurally jaded.