I’m building Qworq - a place to share and build on human–AI conversations
Hi everyone! I’ve been following Product Hunt again after being away from the community for quite a while, so I figured I should finally introduce myself.
I’m Jonathan. I have a writing background and some technical experience, and over the last year or so I’ve spent a lot of time thinking about how people actually use AI — especially the conversations that turn into something unexpectedly useful, strange, creative, practical, or revealing.
That led me to build an early prototype called Qworq.
The basic idea is that the prompt often isn’t the whole artifact. Sometimes the valuable thing is the complete human-AI exchange: the framing, follow-up, correction, judgment, domain knowledge, and unexpected connections that produced the result.
Qworq is an attempt to give those conversations a place to live — searchable, structured, credited, and eventually something other people can build on.
I have a working prototype and a founding collection of 30+ conversations at qworq.io. Right now I’m much more interested in learning than “launching.”
The thing that would help me most is actually pretty simple: if you’re building with AI and have had a conversation that made you think, “this was unusually useful / weird / insightful / worth keeping,” I’d love to know whether Qworq feels like a natural place to contribute it.
Some of the founding examples come from my own interests and experiences, but the scope is intentionally broad — technical, practical, funny, strange, creative, professional, deeply niche, or literally anything else where the exchange produced something worth sharing.
You don’t need to create anything new. The conversation already exists — copy, paste, contribute. Over time, I want to make contributing even easier through direct sharing and browser extensions.
I’m also very interested in hearing where the concept is confusing or where you think I’m getting the behavior wrong.
Glad to be back around Product Hunt and curious to see what other people are building.
p/runevr
p/reso
p/producthunt
p/google-cloud-platform
p/many-displays
p/slotify-2
p/airuncode
p/ai-context-flow
p/kodlo
p/aperture-7
p/raycast
p/lasso-11
p/dino-kids
p/kuc-kuc
p/supabase
p/jev
p/offiai
p/murror
p/dex-by-exmergo
p/htmldrag
Replies
Welcome back, Jonathan. The idea that the correction is part of the artifact matches what I've seen. When I just point out a mistake to a model, the same mistake comes back later wearing a different face. It only stopped once I wrote down the rule behind the correction. So yes, the back and forth often matters more than the prompt.
@hisashispace Hey Hisashi, thanks for the response. You have a fascinating suite of apps and approach to personal dev branding.
It's funny: what you described above was really the genesis of Qworq. I had an extraordinary back-and-forth with a model and it kept using a word inappropriate to the topic of discussion (Spanish flamenco music). I kept pointing out the error but the model became almost 'passive-aggressive' and would omit the word and then bring it back later and then apologize (but use the word AGAIN in the apology!) and so on...
This was so odd that I sought out a place to post it - but was surprised to find that there was no dedicated platform online to post weird, funny, creative, intriguing, insightful, or even practical and technical exchanges. So I decided to create it myself (which became Qworq).
As an aside - Kagemusha was the first Jidaigeki film I ever saw (when I was 11 years old!). Now, many years later I am still a fan and have a vast collection (100+) of 1960s/70s Jidaigeki films on DVD. It's amazing how unknown all that incredible talent from that era is outside of Japan: actors like Tatsuya Nakadai, Tomisaburo Wakayama, Shintaro Katsu.... directors like Hiroshi Inagaki and Masaki Kobayashi, etc.