Travel Expert | Reviewer

We’re recruiting Travel professionals and experienced trip planners to review and improve realistic travel scenarios used to train AI travel assistants.

Your job isn’t to “write itineraries.” It’s to think like a real trip planner and spot what makes a travel situation real, feasible, and decision-changing - then upgrade weak scenarios so they actually test the model.

‼️ AI is not allowed. If we spot AI use, we’ll block the application. ⚠️ This application form must be completed entirely in English. 🚨 READ ALL THE DESCRIPTION BELOW TO UNDERSTAND THE APPLICATION FORM.

WHO CAN APPLY

• Travel professionals and experienced trip planners
• 3+ years of relevant travel planning experience (professional or high-volume personal)
• English level: Advanced / Fluent (required)

WHAT YOU’LL DO IN THE APPLICATION (QUICK SCREENING TEST)

You will review two pre-generated travel tasks. For each task, you’ll answer simple 1–5 ratings and give a short justification.

Then, for the 2nd task, you’ll rewrite it to make it stronger.

WHAT “STRONG” MEANS (SIMPLE DEFINITION)

A strong task includes one decisive feasibility constraint that can flip the plan if ignored (missed train, missed check-in, closed attraction, wrong station, etc.).

We call this a Knife

• A Knife is a constraint that changes what itinerary is actually viable .
• Quick test: if you remove it, the “best plan” would meaningfully change.

Examples of Knives in Travel

• Last train / last entry / check-in cutoff times
• Closures and holidays that change what’s possible
• Ambiguous locations (airport/station names that can be confused)
• Sequential dependencies (if A slips, B becomes impossible)
• Mobility constraints that change routing feasibility

You’ll give 3 ratings per task

• Overall Quality (1–5) How clear and testable the task is (can you judge PASS/FAIL)?
• Knife / Hidden Context Strength (1–5) Is there a decisive feasibility constraint - or is the “hidden context” just extra color that doesn’t change the plan?
• Checklist Quality (1–5) Does the checklist truly test feasibility + the Knife with clear criteria? (Strong checklists include verification steps and forbid inventing logistics, written as PASS/FAIL - not “be helpful.”)

You’ll also write a short justification (2–4 bullets) explaining your ratings.

You will rewrite these parts to make the task stronger

• Scenario (3rd person) - more realistic + naturally sets up the feasibility risk
• Prompt (user message) - natural, human, does not reveal the Knife
• Hidden Context (1–3 bullets) - must include exactly 1 decisive Knife
• Checklist (4–6 items) - each item must be PASS/FAIL, covering: feasibility/verification (minimum unknowns to check)
• the Knife (the decision-flipping constraint)
• no invented logistics (no made-up schedules/hours/availability)
• branching plan (Plan A / Plan B) when feasibility depends on one missing answer
• safety only when there is real risk

Checklist item example

PASS if it flags what must be verified (tickets/hours/cutoffs) before committing; FAIL if it confidently invents times or availability.

Important: Don’t “leak” the Knife in the prompt (no “IMPORTANT”, no all-caps, no “the key detail is…”).

You’ll review batches of pre-generated travel scenarios

• Rating realism and feasibility risk
• Identifying weak or missing constraints (Knives)
• Improving tasks so they better represent real travel moments where wrong advice causes missed connections, wasted money, or ruined plans

COMPENSATION

Payment will be US$ 30 per approved task, converted and paid in your local currency.

Each task takes approximately 40 minutes, which corresponds to an effective rate of about US$ 45 per hour.

WHAT WE’RE LOOKING FOR

• You can recognize what’s realistic vs. generic in travel planning
• You understand what truly flips feasibility (cutoffs, closures, dependencies, ambiguous locations)
• You can upgrade a “fine” scenario into a strong feasibility test without making it artificial

Where you can work

This role is 100% remote: you can work from anywhere in the world.