TechPrufer
Get started
Airbnb
Airbnb logoAirbnb · Feature trail

Guest Journey Sequence Modeling for Search Personalization

2 posts · Jul 2026 – Sep 2026

What this trail covers

This post introduces Chronon's Near-real-time (NRT) Model Transform and Push Mode capabilities, enabling the guest journey sequence recommender to update embeddings in near real-time (under a minute) instead of daily batch runs. This significantly reduces data staleness, leading to improved search result relevance and a measurable increase in uncancelled bookings. These platform capabilities are open-sourced and designed to be general-purpose for other near-real-time modeling efforts.

All Airbnb trails
  1. The guest journey, updated in real time: extending Airbnb’s sequence recommender with ChrononIntroduced Push Mode and Near-real-time (NRT) Model Transform capabilities to Chronon. Push Mode enables event-driven notifications for new feature values, while NRT Model Transform allows model inference (specifically the sequence encoder) to run within the streaming pipeline. This integration reduces embedding staleness from ~2 days to under a minute, improving search ranking quality by +1.67% NDCG and resulting in a ~0.33% increase in uncancelled bookings.
  2. Personalizing Airbnb search by learning from the guest journeyThis post details the implementation of a Transformer-based sequence modeling system for Airbnb search personalization. It addresses challenges of modeling long and sparse event sequences by splitting them into long-term (bookings, reviews, cancellations) and short-term (listing views) components. Efficiency strategies for training include batching of searches, sequence length bucketing, and sparse calculation of searches. Serving efficiency is achieved by decoupling inference into a daily batch job for sequence encoding and real-time retrieval for ranking. The introduction of a setwise ranker, co-trained with the sequence encoder, further enhances personalization by considering candidate listings collectively.