Company profile · 3 min read
Stardrift: AI travel planning with a visible itinerary output
Stardrift is an AI travel planner for finding and booking flights, hotels and activities, with public itinerary examples.
Cofounder of Lazyweb (a16z Speedrun). Previously grew subscriptions and built native ads at Duolingo. Writes First1000 (90,000+ subscribers).
Published · Updated
Stardrift is an AI travel planner with a public example of the output it wants to create.
What it does
The YC profile describes Stardrift as an AI travel planner that helps users find and book flights, hotels and activities. The current site is spare, but it links to a trip-planner path and a public example: a three-day London itinerary with kids.
That example matters more than a generic “AI travel” headline. A useful travel product needs to handle constraints, preferences, sequencing and the difference between research and booking. The buyer should be able to see what the assistant produces before handing over a real trip.
Why I’d look closer
The founder background is unusually relevant to the technical side. Leila Clark is described as a former Jane Street software engineer with a Princeton computer-science degree. A travel planner still needs more than a strong model, though: current inventory, cancellation rules, price changes, handoff and payment boundaries matter.
The public pages do not publish pricing in the checked packet. They also do not show enough detail to assess the booking path itself. That is not a criticism of the idea; it is the next buyer question.
What I’d ask
I would ask whether Stardrift is primarily a research layer or a booking agent, which suppliers are covered, how it handles stale prices, and what the traveler must confirm before anything is purchased. I would also test a constrained trip with children, accessibility needs or a strict budget before trusting a polished itinerary.
My editorial take
Shortlist Stardrift if planning is the painful part of travel and you want a conversational starting point. Treat it as a research assistant until the public booking, price and cancellation boundaries are clearer. The public itinerary example is the right instinct: show the work before asking for trust.
Quick facts
| Field | Sourced detail |
|---|---|
| Product | AI travel planning for flights, hotels and activities |
| Buyer | Travelers who want itinerary research and booking help |
| Pricing | Not published in the checked pages |
| Public evidence | Public trip-planning example and company description |
| Main question | Where does planning end and purchase authority begin? |
Sources checked
| Source | Checked |
|---|---|
| YC company profile | 2026-09-19 |
| Stardrift trip planner | 2026-09-19 |
| London itinerary example | 2026-09-19 |
Cohort context
Stardrift is listed in Summer 2024. In our 2026-09-18 directory snapshot, 18 of 248 listed companies in that cohort have YC’s primary industry label Consumer (7.3%). This is a current-directory comparison, not an original intake count or a performance ranking. Nine-cohort dataset.
Public website snapshot
Observed 2026-09-19T16:17:41.937Z in raw homepage HTML. This records visible metadata and advertised links, not agent execution or product quality.
| Signal | Homepage observation |
|---|---|
| Product description metadata | Observed |
| Canonical link | Observed |
| H1 or H2 heading | Observed |
| Typed structured data | Observed |
| Docs/developer link | Not observed in this response |
| Pricing link | Not observed in this response |
| llms.txt link | Not observed in this response |
| Markdown alternate | Not observed in this response |
Public observations · Collection method. Missing links here do not establish that a capability or file is absent elsewhere.
About the author
Cofounder of Lazyweb (a16z Speedrun). Previously grew subscriptions and built native ads at Duolingo. Writes First1000 (90,000+ subscribers).
I cofound Lazyweb and publish Mudpie. This is an owner-written publication, not an independent testing organization. Research notes distinguish observations, sourced reporting and editorial judgment.
First1000 ↗ · X ↗