AI-powered flight discovery — built with early users, not just for them
We’re building an AI-first way to search and discover flights — focused on simplifying decisions, not overwhelming you with options.
But here’s the truth: This is still early.
Instead of waiting for a “perfect” launch, we’re opening access to a small group of early users who want to help shape how this product evolves.
If you’ve ever:
Spent hours comparing flights across tabs
Felt unsure if you’re actually getting the “best” option
Wanted a smarter, more intuitive way to search
This might be worth trying.
#What it does (so far)
AI-assisted flight recommendations based on your preferences
Cleaner, less cluttered experience compared to traditional booking sites
Designed to simplify decision-making, not complicate it
#What to expect
This is a beta experience — things may break or feel incomplete
Recommendations are improving continuously
Some flows are still being refined
# Who this is for
We’re specifically looking for people who:
Are planning a trip soon (next 30–60 days)
Book flights frequently
Are willing to share honest, direct feedback
# Why join now
Early users will:
Get direct access to the product
Influence how features evolve
Help shape the core AI experience
This isn’t just early access — it’s co-building with us.
#Access
We’re onboarding a limited number of users in small batches.
If you’re interested, drop a comment or sign up — we’ll reach out with access.

Replies
Love this approach to flight discovery...building with early users is the way to go! Sounds perfect for my next trip planning session...I'll check it out.
If you're up for it, I'm launching The Sponge on PH soon...an AI-powered flashcard app that turns webpages into study material with spaced repetition. Would appreciate a follow (PRODUCT HUNT LAUNCH link in my profile).
@rianbrob I am glad to your interest into this project! Check out bookwithai.in once and try getting most out of this platform and share your feedback for improvisation (Feedback button is in Bottom right corner).
@george_adriel Co-building approach is smart, especially for something as preference-heavy as flight search. Curious though, how is the AI actually making recommendations right now? Is it learning from my past choices, or is it more prompt-based where I describe what I want? That distinction changes the experience a lot.
@vishal7017 Great question — and you’re absolutely right, that distinction changes the experience a lot.
Right now, it’s primarily prompt + preference-driven.
Users describe what they care about (price, duration, number of stops, timings, etc.), and the system tries to prioritize results accordingly and simplify the decision.
We’re not heavily relying on past behavior yet, mainly because in early stages we want to:
Understand explicit user intent clearly
Avoid making incorrect assumptions too early
That said, moving toward a hybrid model is definitely the direction:
Short term gives better interpretation of what users ask for
Longer term gives learning from patterns, repeat behavior, and preferences over time
One thing we’re actively exploring is:
How much automation users actually want vs how much control they want to retain — early feedback has been quite split here.
Curious from your side — would you trust a system that auto-learns your preferences over time, or would you prefer staying in control each time you search?