Show HN: Roundtable – A rubber duck that argues with itself(roundtable.ovlo.ai)
roundtable.ovlo.ai
Show HN: Roundtable – A rubber duck that argues with itself
https://roundtable.ovlo.ai
4 comments
Tried it out, pretty neat. Better than the positive affirmation from ChatGPT when ideating at least.
The personas were brutal in the best possible way. Great job
Love this. I’ve had the same issue using LLMs to sanity-check product decisions - they’re too agreeable if you give them any context that reflects your own thinking. Also even if you ask them to be brutal they sort of always come up with some bs Hail Mary positive spin. I've been playing around with this for best in practice QC doc (ironically) and the internal debate is pretty helpful. I will keep you posted on how it goes though. Would be good to be able to add more than 2 personas though...
Super valid, only really added the personas I was using for my own use case but we added a create your own persona ability. even made it with use ai so it wont take too much time!
Except it was obvious to my cofounder I wasn't really validating anything. LLMs are incredibly good at agreeing with you in subtle ways, especially when you feed them context that already reflects your thoughts. I'd ask "Does this make sense?" and get a beautifully worded essay about why yes, obviously, this is the best thing ever.
I was using AI as an echo chamber without noticing. It was like a game, you bias the model a little, it biases you a little back.
So my cofounder hacked together a tiny internal tool over the weekend to break that loop. Instead of one model validating my thinking, it puts multiple personas with different expertise in the same conversation. They naturally push back on each other.
It's basically a rubber duck that argues with itself.
We gave it some UI and called it Roundtable: https://roundtable.ovlo.ai
Would love to hear what you'all think!