Queries for "night view" missed "scenic evenings." AI now matches intent across 1.2M properties in <100ms, regardless of phrasing.
One of the world’s largest online travel service providers offers access to over 1.2 million accommodation choices for travelers across more than 200 countries.
Legacy keyword search failed to capture intent, meaning queries for "beautiful night view" missed relevant listings described as having a "scenic...
Online travel service provider for lodging, transportation, and packaged tours.
Vector database platform for building and scaling AI applications.
Trip.com Group's Hotel search is part of this use case:
Related implementations across industries and use cases
Elasticsearch kept every vector in RAM—cost-prohibitive at scale. One shared Milvus platform now runs 100+ workloads on 10B+ vectors.
Every data question routed through engineers. Now 800+ users — 70% non-technical — query a 200-country travel dataset in plain language.
Keyword search failed on complex vehicle queries. Now, an AI pipeline rewrites messy questions to power shopping and service assistants.
Keyword search failed on complex vehicle queries. Now, an AI pipeline rewrites messy questions to power shopping and service assistants.
Scattered data made AI features impossible to build. Unifying it unlocked search that understands traveler intent, not just keywords.
Manual prospecting across multiple tools delayed first contact. Now, an agent curates warm leads, so reps start their day ready to sell.
Endless phone tag bottlenecked 10,000 yearly hires. Now, AI screens and schedules via text, freeing managers for in-person interviews.
Tournaments running simultaneously meant an hour of manual checks each. AI agents now run them in minutes, freeing the team to be proactive.
Large AI training jobs meant fighting for preemptible slots or leaving campus. Marlowe gave any lab guaranteed multi-node access on demand.
Queries for "night view" missed "scenic evenings." AI now matches intent across 1.2M properties in <100ms, regardless of phrasing.
One of the world’s largest online travel service providers offers access to over 1.2 million accommodation choices for travelers across more than 200 countries.
Legacy keyword search failed to capture intent, meaning queries for "beautiful night view" missed relevant listings described as having a "scenic...
Online travel service provider for lodging, transportation, and packaged tours.
Vector database platform for building and scaling AI applications.
Trip.com Group's Hotel search is part of this use case:
Related implementations across industries and use cases
Elasticsearch kept every vector in RAM—cost-prohibitive at scale. One shared Milvus platform now runs 100+ workloads on 10B+ vectors.
Every data question routed through engineers. Now 800+ users — 70% non-technical — query a 200-country travel dataset in plain language.
Keyword search failed on complex vehicle queries. Now, an AI pipeline rewrites messy questions to power shopping and service assistants.
Keyword search failed on complex vehicle queries. Now, an AI pipeline rewrites messy questions to power shopping and service assistants.
Scattered data made AI features impossible to build. Unifying it unlocked search that understands traveler intent, not just keywords.
Manual prospecting across multiple tools delayed first contact. Now, an agent curates warm leads, so reps start their day ready to sell.
Endless phone tag bottlenecked 10,000 yearly hires. Now, AI screens and schedules via text, freeing managers for in-person interviews.
Tournaments running simultaneously meant an hour of manual checks each. AI agents now run them in minutes, freeing the team to be proactive.
Large AI training jobs meant fighting for preemptible slots or leaving campus. Marlowe gave any lab guaranteed multi-node access on demand.