CEE Travel is a Central European hospitality technology company focused on modernising accommodation discovery. Our challenge was to help travellers navigate 18,000+ hotel listings through natural language — delivering a curated shortlist in seconds instead of a scroll through hundreds of results.
A leading online travel agency lists over 18,000 accommodations across five Central European countries. For travellers, this creates a paradox of choice: finding the right hotel means wading through hundreds of listings, comparing ratings, reading reviews, and cross-referencing locations. The problem is especially acute for the Hungarian-speaking market, where traditional keyword search cannot capture intent - a query like “romantic weekend with spa, quiet but close to Budapest” is simply beyond what filters can understand.
CEE Travel needed a way to turn 18,000 properties into a personalised shortlist, using the same natural language a traveller would use when asking a friend for a recommendation.
We designed two interconnected systems: an AI pipeline that enriches and indexes all 18,000+ hotels, and a real-time travel agent embedded directly into the leading online travel agency’s website. We focused on:
AI-powered hotel catalogue: We used Amazon Bedrock to generate rich, search-optimised Hungarian-language descriptions for every hotel — capturing location, character, amenities, and surrounding area in a way that makes AI search dramatically more effective than working with raw listing data.
Intent-based search: Amazon OpenSearch understands language, location, and intent simultaneously - so a search for “quiet romantic weekend near Budapest with a spa” returns the right hotels, not just the ones that contain those words.
Conversational travel agent: Amazon Bedrock holds a natural conversation with the traveller - collecting preferences for destination, dates, travel style, and budget - then automatically surfaces the best matches once it has enough context.
Personalised recommendations: Every recommended hotel comes with a personalised explanation - generated by Amazon Bedrock - of why it specifically matches what that traveller described, delivered in real time as results load.
Zero-friction delivery: The AI assistant is embedded directly into the website as a lightweight Chrome extension — no app download, no separate website. Available on the Chrome Web Store.
The extension is live on the Chrome Web Store and available to all users of the leading online travel agency across the CEE region. Early production data shows an 85% reduction in time-to-shortlist - from around 15 minutes of manual browsing to under 2 minutes - and a 3.2x increase in click-through to property detail pages. 18,000+ hotels are now semantically indexed and searchable in Hungarian through natural language, with zero infrastructure management overhead.
CEE Travel is expanding the platform across Croatia, Romania, Slovakia, and the Czech Republic, with a roadmap that includes direct booking API integration, multi-language support, and a white-label B2B version for travel agencies.
Tell us more about yourself and what you're got in mind.
Prefer email? hello@apexlab.io
Or even a quick call?
We don’t care about fancy CVs or long interview processes. Provide some basic information, let’s have a quick chat, and then show us what you love doing the most: write some code and create a product.
