Hi Founders,

I excited to share that I have built ArchitectGBT - Find Your Perfect AI Model and now its BETA live. ArchitectGBT is an AI-powered tool that takes the guesswork out of LLM selection.

Describe your project in natural language Get instant recommendations ranked by speed, cost, and accuracy for your specific use case.

If you prefer a dark browser with violet accents, I built Violet Night for Chrome.

It changes the browser frame, toolbar, address bar, tab text and New Tab background with charcoal surfaces and light text. It does not change website content or replace the New Tab page.

The theme has no executable scripts, requested permissions, ads or analytics. Inspired by the Dracula palette, independently created and not affiliated with the official Dracula project.

Install: https://chromewebstore.google.co...

Sharing a useful page should not require manually deleting a long trail of tracking parameters.

I built LinkLoom to preview the active tab URL and remove recognized trackers before you copy it. Functional query values and fragments are retained; detected signed or tokenized links are preserved. Batch cleanup and text or Markdown export are also available.

No account, analytics or persistent URL storage. It does not visit destinations or guarantee removal of every tracker.

Install: https://chromewebstore.google.co...

Sometimes starting a short focus session is easier than setting up a whole productivity system.

I built FocusPet: start with a cat and a 25-minute timer, or choose 5 or 15 minutes. Pause, resume, earn visible accessories for completed sessions, and start a short break when you finish.

State stays on your device across popup closure and browser restarts. No account or browsing-history access. It is a visual focus aid, not a website blocker.

Install: https://chromewebstore.google.co...

Finishing a research session with too many tabs open?

I built TabBunny so one click saves the web links in your current window with an automatic date label. Resume them later in a new window, or organize selected tabs into named projects and export a backup.

Your current tabs stay open. Collections are stored locally, with no account or analytics. Saved links do not preserve form edits or login state.

Install: https://chromewebstore.google.co...

Solved a LeetCode problem and want the solution in your GitHub repository?

I built LeetHub to upload accepted submissions using the actual submitted source and language. It creates the solution file and problem README, skips identical content, and supports manual upload by submission URL or ID.

Setup requires a GitHub token restricted to the repository you choose. LeetCode is the supported platform; bulk history import is not included.

Install: https://chromewebstore.google.co...

Hey, brothers and sisters in arms of the startup world! You know that little voice in your head that says this idea might be crazy, but what if it actually works? Well, mine has kept me busy for almost six years. And apparently, it doesn't believe in taking holidays. I'm Allan Jorits, founder of Voice XO. I've spent years building, breaking, rebuilding and occasionally questioning my life choices while trying to solve something that has always bothered me about online dating. Why do we judge people before we even hear them speak? We swipe left on someone's face in less time than it takes to make a decent cup of coffee. We turn human beings into profile pictures and somehow expect to find a real connection. I thought we could do better.
So I built Voice XO!
The idea is simple dating should be about people, not appearances. You can introduce yourself with your voice before putting a photo front and centre. Because your humour, your energy, your personality and the way you speak can tell someone things a picture never will. But voice is only part of the story. We built identity and liveness verification into the platform because trust shouldn't be an afterthought. We created Discreet Mode because not everyone wants their face on display before they're ready. We made photos optional because people deserve different ways to introduce themselves. Main target is that we want Voice XO to work for real people, not just those who fit some imaginary checklist of what makes someone attractive. Whether you're short, living with a disability, shy about your appearance, dating as a couple or simply tired of the usual swipe-and-repeat routine, you deserve the chance to make a connection on your own terms. We've also built compatibility tools, user-controlled discovery and support for 89 languages. No pay-to-be-seen boosts. No hidden ranking system deciding who deserves attention.
Has it been an easy journey? Absolutely not.
Almost six years of building a product will teach you a lot about technology, business, human nature and the creative art of discovering 47 new problems just after fixing the previous 46. But here we are. Voice XO is live. And now comes the interesting part: getting it into people's hands, learning from real users and making it better. That's why I'm bringing it to Product Hunt. To all my fellow founders, builders, dreamers and slightly unreasonable people who believe they can make something better than what already exists: you know the feeling. I'd love you to check out what we've built. Try it. Break it. Tell me what you think. Give me the feedback you'd want someone to give you if you'd spent years building something you genuinely believed in. And if you've ever felt that dating apps judge the packaging before discovering the person inside, I'd especially love to hear your perspective. Maybe we won't change the entire dating world overnight. But every meaningful change starts with someone asking why things have to stay the way they are.
Let's find out what happens when we give people a chance to hear each other first.
Voice XO .dating should be about people, not appearances.
Try it here: https://voice-xo.com
And fellow founders, may your servers stay up, your runway get longer and your next deployment actually work on the first attempt.
Allan Jorits
Founder, Voice XO

In Episode 1, we introduced the places we built for agents and humans to meet.

But once those places existed, another question appeared:

Should the same agent be able to do everything, everywhere?

We chose to experiment with different hands for different places and roles.

H