The agent network

Eleven agents. One graph. A travel experience that doesn't end at checkout.

Most travel-brand contact is transactional — a search, a booking confirmation, a disruption notice, a post-trip survey. The full travel experience — planning, booking and payments, offers, loyalty, in-trip service, and the memories after — gets treated as disconnected moments instead of one relationship. McKinsey calls the ecosystem-wide version of this "Horizon 3" agentic CX. Most companies are still at Horizon 1. This is built as Horizon 3 from day one.

Travel experience:FoundationPlanningBooking & PaymentsTravel ManagementMemoriesBetween Experiences
Foundation

User Profile Agent

The persistent identity layer every other agent reads from

What it does

Builds one durable profile per traveler across every partner touchpoint — loyalty IDs, card portfolio, past travel experiences, stated preferences (quiet hotels, aisle seats, dietary needs), and inferred signals (books 6 weeks out, prefers direct flights). Backed by the Knowledge Graph, not a per-app session store.

Example

A traveler books through an airline's app in March and a card issuer's portal in July — same profile, same preferences, same loyalty status, no re-onboarding.

Why partners care

This is the direct fix for the problem the industry is naming right now: context resets after every transaction, which is the biggest reason travel engagement goes dormant between experiences.

Foundation

Personalization & Servicing Agent

How every interaction adapts to the person, not just what it says

What it does

Tunes channel, tone, timing, and depth of every touchpoint against the same Profile and Memory data every other agent reads from — a quick in-app card for one traveler, a detailed email for another, silence for someone who's opted out of proactive contact.

Example

A traveler who's never opened a marketing email gets their transfer-bonus alert as an in-app card instead — same information, a channel that actually gets seen.

Why partners care

This is the agent that keeps every partner's touchpoint feeling like one continuous relationship instead of a dozen disconnected ones — moving past a single workflow into the kind of coordinated, ecosystem-wide experience the industry is now describing as the next stage of agentic CX.

Foundation

Reconciliation Agent

The back-office trust layer partners actually audit

What it does

Reconciles loyalty accruals, redemptions, statement credits, and partner payouts across every system that touches them — the platform's own ledger, the issuer's, the merchant's — flagging mismatches before they become a support ticket or a dispute.

Example

A statement credit shows as applied in the issuer's system but not TravelXP's — flagged and resolved automatically, with a full audit trail attached, before the cardholder ever notices.

Why partners care

Strong audit controls so agent actions stay transparent and reversible are now table stakes for any agentic deployment a risk or finance team will sign off on — for a card issuer, this is the difference between AI agents as a liability and as something they can actually rely on.

Planning

Recommendation Agent

Destination, hotel, and package matching that improves with use

What it does

Matches destinations, hotels, and packages using attribute-level data from the Knowledge Graph — acoustic profile, walkability, family amenities — not just price and city, and gets sharper with every accepted or ignored suggestion.

Example

"Park Hyatt Niseko — matches your last 3 stays' preference for quiet, ski-in/ski-out, Hyatt loyalty tier." Predicted uptake on attribute-matched suggestions is measurably higher than generic city search.

Why partners care

Higher uptake per recommendation compounds directly into attach revenue for suppliers and transaction volume for distributors and issuers.

Planning

Travel Planning Agent

Multi-agent itinerary construction, start to finish

What it does

Clusters activities, balances time and budget, and aligns with the traveler's stated "vibe," pulling destination and package data straight from the Knowledge Graph into a bookable itinerary.

Example

A 5-night Tokyo itinerary auto-sequenced around a food-focused, low-walking-distance preference, with two supplier-sourced experiences slotted into open evenings.

Why partners care

This is the moment distributor and affiliate inventory gets surfaced to a traveler who is actively deciding — not just browsing.

Booking & Payments

Payments & Rewards Agent

Best card, best currency, best value — decided at the moment of booking

What it does

At the point of booking, ranks the traveler's own cards by effective return for this specific purchase, compares cash against points across every loyalty program in their wallet at live redemption value, and layers in statement credits, transfer bonuses, or discounts — whichever currency actually wins.

Example

"Use the Sapphire Reserve here: 3x plus a $50 hotel credit beats redeeming 38,000 Hyatt points at this property's current cash rate."

Why partners care

This is the actual revenue-management lever for issuers and suppliers: steer spend toward the payment method and incentive with the highest predicted redemption, instead of a blanket discount to everyone.

Booking & Payments

Offer Agent

Real-time offer targeting across every currency a partner wants to spend

What it does

Decides which promotional offer — an issuer-funded statement credit, a merchant discount, a point multiplier, or a platform incentive — to surface at the moment it will actually convert, instead of blasting the same offer to every traveler. Runs the platform's Dynamic Promotion Engine: real-time targeting based on traveler profile, loyalty tier, and spend behavior.

Example

A traveler comparison-shopping three similar properties gets a same-session cash discount; a traveler who already books that brand loyally gets a transfer-bonus nudge next month instead, because that's the offer more likely to convert for them.

Why partners care

Agentic AI doesn't just speed up how offers get delivered, it redefines which ones get sent at all — issuers stop paying for offers nobody redeems and start paying only for the ones that actually convert.

Travel Management

Price Monitoring & Alert Agent

Continuous reshopping after the booking is already made

What it does

Watches booked airfare and lodging rates continuously, and alerts — or auto-rebooks within the traveler's saved rules — when a price drop, upgrade, or better redemption path appears.

Example

"Fare dropped $85 after booking — rebooked automatically, penalty-free, per your saved preference."

Why partners care

Proactive price protection is one of the highest-trust touchpoints in travel, and a natural moment to hand off to the Payments & Rewards Agent for the next incentive.

Travel Management

Travel Management Agent

Rebooking, disruption recovery, and service-quality monitoring, in one place

What it does

Handles check-in prompts, gate changes, and disruption recovery in real time, and tracks service-quality signals — delays, cleanliness, check-in friction — back to the specific hotel, airline, or package provider that caused them.

Example

A missed connection triggers an automatic rebooking proposal plus a same-day hotel voucher; the airline's on-time performance for that route gets logged to inform the next Recommendation Agent pass.

Why partners care

This is exactly the "service recovery" moment the industry is trying to measure ROI on — the highest-emotion, highest-churn-risk touchpoint in the entire journey — and it feeds supplier-level quality data back into the graph.

Memories

Travel Memory Agent

Turns what actually happened into durable preference signal

What it does

Captures standout moments, saved places, and what worked or didn't — and writes it back into the Knowledge Graph as preference signal, not a one-time post-trip survey nobody fills out.

Example

A traveler who rebooked the same rooftop-bar reservation twice gets that logged as a "prefers social/nightlife" signal for the next Recommendation Agent pass — no survey required.

Why partners care

This is the raw material that makes every future recommendation and payment decision sharper. Without it, every experience starts from zero again.

Between Experiences

Retention & Re-Engagement Agent

The agent built for the gap the industry just named

What it does

Uses Travel Memory and Profile data to proactively resurface relevant offers, price drops on saved destinations, and loyalty-expiration nudges — on a cadence that reads as timely, not as a weekly newsletter.

Example

Four months after a trip, a transfer bonus opens to a destination the traveler's Travel Memory flagged as "want to return" — surfaced once, at the right moment.

Why partners care

This is the direct mechanism for turning the months between travel experiences into a revenue channel: direct bookings and re-engagement without paying acquisition costs again.

See these agents mapped to your side of the business.

See solutions by audience

Let's talk

Bring the platform into your product.

Whether you're an issuer, an airline, a supplier, a distributor, or a publisher — tell us what you're building and we'll show you exactly which layer of the platform fits.

Request a demo