Elisha Ong

Zesty by DoorDash - Your personal restaurant concierge

by
The Problem: Scrolling map pins and articles is exhausting and star ratings don’t capture a "vibe". The Solution: Zesty is an AI concierge that turns social signals and TikTok trends into the perfect meal. Conversational AI: Chat like a local friend for hyper-specific spots (e.g., "cozy pasta with low light"). Vibe-First: Discover by noise level, lighting, and hype. Personalized: It learns your taste to find what you’ll actually love.

Add a comment

Replies

Best
Gianmarco Carrieri

The trip planning feature is the most interesting unlock here — restaurant discovery is usually the hardest part of a trip to plan well, because it depends on vibe, neighborhood context, who you're with, and how meals sequence across the day. Most travel tools treat food as an afterthought (or a list of Yelp stars). The fact that Zesty can reason about day-by-day food itineraries changes the mental model entirely. Building something adjacent at Aitinery for travel — the challenge we keep running into is the gap between "recommend a restaurant" (solvable) and "sequence meals across a 10-day trip so they feel intentional and varied, not just highest-rated" (much harder). Does the trip planning mode handle temporal sequencing across days, or is it more discovery per stop?

Brad Willard

@giammbo This is one of our favorite capabilities and I'm so glad you've discovered it. I would definitely suggest trying to see how far you can push it. You can also ask Zesty to make lists for you. "I'm going to Toyko for 10 days, make me a list of all the top coffee places to try."

Gianmarco Carrieri

The Tokyo coffee list example makes the distinction clearer — that's aggregation, not sequencing: AI can rank and filter within a single category. The harder mode I was probing is cross-constraint: day 3 lunch should be near the museum, different cuisine from last night, and ideally leave room for a dessert spot close by. Does the trip planning feature try to hold context across days, or is each stop still a fresh query?

Elisha Ong

@giammbo totally, this has been one of my fav ahah moments using it to plan trips in areas i'm completely new too, or suggesting some things to do for the night out!

Elisha Ong
@giammbo I’m curious how much you can push Zesty and see how much context it’s able to preserve or recommend
Gianmarco Carrieri

Going to properly stress-test it — will take it through a full multi-day trip with real constraints (different cuisine each night, proximity to where I'll actually be, no repeats). That's the use case I care most about. Thanks for building something worth pushing.

Nika

This app should claim Gordon Ramsay as an ambassador :D

Elisha Ong

@busmark_w_nika haha totally! great idea

Gianmarco Carrieri

The 'new to an area' framing is actually the strongest use case — it's where AI has no competition from accumulated local knowledge, so there's no shortcut alternative. The night-out dimension sounds distinct: spontaneous + social context vs. the more deliberate trip-planning mode. Do you find users tend to anchor on one or the other, or do they move between both depending on the moment?

Nika

@elishaong Next challenge unlocked :D

Elisha Ong
@busmark_w_nika totally 😂
Elisha Ong

Hey Product Hunt! 👋

We’re so excited to share Zesty with you all today. 🍋

We’ve just rolled out a major refresh that makes discovery feel less like a chore and more like a conversation. Thanks to recent agentic model improvements, Zesty has gotten a lot smarter—it doesn't just search; it reasons with you to find the exact vibe you’re after.

As our co-founder Andy Fang mentioned in his post, we’re having a ton of fun playing with these agentic capabilities, and we’re looking forward to incorporating this intelligence into the broader DoorDash ecosystem soon.

What can you do with Zesty today?

  • Vibe Search: Talk to it like a local friend (e.g., "Find a moody wine bar near Industry City").

  • Multiplayer & Personalization: @ your friends to combine your tastes and see what they’re loving.

  • Trip Planning & Global Search: Planning a trip? Use it globally to build day-by-day food itineraries.

  • Lists & Social: Create "food playlists" and follow others for real word-of-mouth recs.

  • Reservations: Check availability and book a table directly without app-hopping.

The latest version is now available on the App Store! Check out our X account @getzestyapp to see the AI in action.

We’re live all day—drop your craziest "vibe" prompt in the comments and let’s see what Zesty finds for you! 🍝🚀

Abhinav Ramesh

So useful, especially the vibe search part! Is it available across the globe?

Brad Willard

@abhinavramesh Yes it does work anywhere in the world. The underlying system gets smarter over time as well. So the more people search in different areas the better it will get in those areas. For this reason, even though it does work anywhere, it's best in major cities.

Gabriel Savit

Congrats Manolo and team! Excited to see where this goes.

manolo

@gs_ thank you!! we're very excited to launch and see what everyone thinks!

Victor N

Congrats on your launch, would you please explain me what would be the difference between using Zesty and providing some prompts to a publicly available AI tool that maybe has already a history of interaction with me and knows my tastes. Also, did you put geographical limitations for the app store? I was not able to find it on apple store on my mobile, I searched using the exact same name " Zesty, zesty; Zesty - Local Food Discovery" . I wanted to try it out myself but was not able to do so.

Brad Willard

@viktorgems It's only available in US and Canada app stores. Perhaps that's your issue? Zesty is different in a few key ways and I encourage you to side by side our recommendations to more generic AI tools, especially for complex multi intent queries. A lot of existing AI tools are mostly doing a few web searches to answer your question. Zesty sits atop a very specialized agentic search stack that's populated by complex deep research style llms hunting many sources for very nuanced information. This enables Zesty to be way more nuanced, granular and specific. This also means as it learns your preferences those are also way more nuanced, granular and specific.

Shan

Looks cool! Is there an android/web app?

Brad Willard

@shansingh Not yet, but it's on our radar. We want to get as much feedback as possible and address that first before expanding.

Elisha Ong

@shansingh we do have a shareable web link to chats when you share from the app. For example "What are Alysa Liu’s fav restaurants in sf Bay Area" https://www.zesty.com/v1/share/conversation/1620924b-20d3-4400-a75b-186b4aac50d0

Sarim Haq

Amazing work team!

Elisha Ong

@sarim_haq thank you! What's your favorite feature or prompt so far?

Sarim Haq

@elishaong I like that it pulls in reviews from all mutliple sources not just Google Maps!

Emmanuella Udoh

The "cozy pasta with low light" example got me because I've literally typed something like that into Google before and found nothing useful. I think what Zesty is doing differently is it's not trying to give you more options, it's trying to give you the right one. Noise level, lighting, vibe — that's how we actually describe places to our friends, and no app has really figured that out yet.

My only question is around the TikTok side of things; what happens when a spot goes viral and suddenly loses the vibe that made it worth recommending in the first place? Would love to know how you're thinking about that.

But honestly, really cool launch. The concept feels fresh without being overdone. Excited to try it

Elisha Ong
@emmanuella_udoh thanks for the note and yes that’s exactly the problem we’re trying to solve and where we think LLMs do a much better job than traditional search. The content is constantly refreshed… so if a place that used to trend and taper off, Zesty should theoretically know about it too (I hope! 🙏)
Mykyta Semenov 🇺🇦🇳🇱

A really cool idea) We’re building a similar service, but for tourism. We’ll be launching on Product Hunt this month)

12
Next
Last