Steven Hacker

Sentence rhythm turned out to be a stronger voice signal than vocabulary how long someone runs their sentences, where they hedge versus assert. Early drafts were accurate but felt weirdly formal. Adjusting syntax patterns made a bigger difference than swapping words.

The most interesting engineering insight from the last six months: sentence rhythm is a stronger voice signal than vocabulary. Two people can use the same words and sound completely different. Getting rhythm right is what takes a persona from "close" to "that's actually them."

Hey Product Hunt I'm Steven Hacker (yes, that's actually my last name).

I came out of the film industry, where the whole job is making a voice feel real getting a character to sound like a person, not like someone reading lines. You spend years obsessing over what makes speech feel authentic versus fake.

That obsession is what pulled me into building MePersona. Creators today are drowning in fan messages across a dozen platforms, and the "AI reply" tools that promise to help all have the same flaw the fan can tell it wasn't really them. Coming from film, that gap is the only thing I care about: a reply that doesn't sound like you is worse than no reply at all.

So I'm building the tool I think should exist one that learns how a creator actually talks and drafts in that voice, everywhere they get messages.