Antony Shelcott

We built an AI on WhatsApp to cut through the dining noise. Here's what we learned

by

We noticed something that kept bothering us: finding the right place to eat had somehow gotten harder, not easier.

Every recommendation platform had quietly been gamed, restaurants paying for placement, reviews written by agencies, "best of" lists recycled year after year. Locals knew it. Visitors just got burned.

So instead of building another app, we built Yenta, an AI friend on WhatsApp. You text your vibe ("rooftop, something buzzing, not a tourist trap"), and it responds like a friend who actually knows the city.

We go live May 15th. Waitlist is open at https://tr.ee/gOPXT4QG7R

A couple of things we're genuinely wrestling with — would love to hear from this community:

  1. How do you find restaurants in a new city, or even your own? Apps, friends, doomscrolling, what actually works?

  2. For anyone who's built recommendation products: how did you handle the cold start problem when personalisation data is thin? We're curious how others navigated "trust before you have enough signal."

Happy to talk through anything, the product, the market, or why we put it on WhatsApp of all places.

7 views

Add a comment

Replies

Best
Rian Robertson

Neat approach sidestepping the app store noise with WhatsApp! Love it. For cold starts, I've seen teams bootstrap with hyper-local expert seeding before personalizing. How are you tackling trust early on?

If you're up for it, I'm launching The Sponge...an AI-powered flashcard app that turns webpages into study material with spaced repetition...on PH soon. Would appreciate a follow (See "PRODUCT HUNT LAUNCH" Link in my profile)