Guests now expect WhatsApp-speed answers, on every channel, in every language. The bar has moved.
The AI orchestration layerfor hotel operations.
Resolve guest requests, execute across your PMS and operations stack, and run on every channel — one orchestration layer trusted by 60+ European hotel groups.
Hotel operations map
Skills Engine · live routing
Available for messaging, phone and email
D3x Skills Engine
Skills mapped to every layer of hotel operations.
Front Desk
Concierge
Back office
Housekeeping
Sales
Marketing
Trusted by leading hotel groups
IN PRODUCTION
+100K
messages per month
120x
faster median response (15m → 7s)
60%
containment · zero human touch
+95%
volume growth · no headcount increase
Same metrics, every quarter. Not a roadmap.
THE MOMENT
The category-defining vendor will be chosen in the next 18 months.
Decision window
18
months
Buyer selection is happening now, in real procurement processes, on multi-year contracts.
Generic AI cannot execute against your PMS, housekeeping, table-management, or CRM. The work is in the integrations.
The hotel groups deploying now are setting the operational baseline competitors will spend years catching up to.
BUILT FOR WHAT'S NEXT
Your stack for the next era of hospitality AI.
The groups winning the next procurement cycle are not buying another inbox tool — they are standardising on a platform that compounds. Hospitality AI is moving from chatbots to systems with a knowledge base that self-heals and self-learns, MCP-native access for the copilots your teams already use, agentic email that executes governed back-office workflows, and an open orchestration layer that connects your stack — not replaces it. D3x is engineered for that shift.
KNOWLEDGE BASE
The knowledge base of the future
Self-healing and self-learning by design — not a static FAQ that goes stale every season. Every unresolved question, low-confidence answer, escalation, and staff correction becomes structured signal. D3x diagnoses the root cause, drafts the smallest safe update, and routes it through human review before anything changes in production.
- Self-healing: closes gaps surfaced in real guest conversations
- Self-learning: improves from operator-approved diffs — not blind retraining
- Find conflicts between human message and the knowledge base
knowledge.d3x
Self-healing loopCapture escalations, corrections, and retrieval gaps
Diagnose missing policy vs. stale source vs. PMS conflict
Propose the smallest safe knowledge diff
Operator approves — then production updates
OPEN PROTOCOL
Plug Claude, Codex, and your copilots into D3x
Model Context Protocol is the open standard for connecting AI applications to real tools and data — one integration that works across MCP-compatible hosts. Wire D3x into Claude Desktop, Cursor, internal copilots, or third-party operators: same agents, Skills Engine, guardrails, and audit trail, wherever your technology team already works.
- Connect once — operate from any MCP-compatible LLM client
- Pull live production reports and run ops without opening the dashboard
- Scoped credentials, tool policies, and audit on every call
D3x MCP
Model Context Protocol
PMS · ops · BI · guest agents
EMAIL AGENTS
Agentic email for hotel back office
Full-thread autonomy — not a copilot that drafts one line. D3x reads guest and finance threads in Gmail or Outlook, authenticates the sender, matches the reservation, applies your policy graph, executes booking and invoice workflows, and routes sensitive actions through staff Agree before anything is sent.
- Entire thread in context — guest, finance, and corporate senders
- Governed execution: booking changes, invoice updates, folio actions
- Human Agree before commit — same Skills Engine and audit as messaging
Inbox
Resolution readyInvoice name change — BK-48291
Sarah Chen
Please change the name on my invoice to Chen Consulting Ltd. Booking BK-48291.
Intent: invoice modification
Guest authenticated · folio linked
Billing name updated · PDF regenerated
OPEN PLATFORM
A pipe through your stack — not a walled garden
D3x is the AI orchestration layer — not another all-in-one trying to displace your PMS, CRM, or ops tools. We connect openly to the systems you already run, treat integration partners as first-class collaborators, and build lasting relationships across the hospitality ecosystem so you keep best-of-breed flexibility without gridlock.
- Orchestration above your stack — no rip-and-replace of systems of record
- 25+ live integrations · new connectors in ~3 weeks with API access
- Partner-friendly by design — open APIs, shared roadmaps, ecosystem relationships
integrations.d3x
Open platformYour stack
PMS · Housekeeping · CRM
D3x orchestration
Guest intent → routed action
Systems & partners
APIs · integrations · audit
AGENTS
Every channel. Every guest. Every action.
Three production agents — each with the same Skills Engine, Knowledge Base and integration fabric underneath. Every guest is authenticated and matched to their reservation before the agent acts. See what guests experience and what runs in the background.
WhatsApp, OTA inbox, web chat, SMS, and social DMs — one agent per property. D3x authenticates every guest and matches the conversation to the right reservation in your PMS before acting.
70–85% autonomous resolution
What the guest sees
WhatsApp · Riverside Hotel
Typically replies in seconds
Hi, can I book a table in the hotel restaurant for dinner tonight? We're two people.
12:01
Yes. I can reserve Riverside Restaurant at 7:30pm for two. Shall I confirm that?
12:01
Perfect, please book it.
12:02
Done. Table for two at 7:30pm is confirmed.
12:02
Also, could we check out later tomorrow? Is that possible?
12:04
We can extend checkout to 2pm tomorrow for €35. Shall I confirm that and add the charge to your room?
12:04
Yes please, that works.
12:05
Done. Late checkout until 2pm is confirmed. €35 has been added to your room bill. Enjoy dinner!
12:05
What D3x runs
12:01:01
inbound.whatsapp.received
Intent: restaurant reservation · tonight
12:01:01
guest.authenticate
WhatsApp number + property → Sarah Chen verified
12:01:02
reservation.match
Thread linked to reservation · room 412 · in-house
12:01:03
restaurant.availability.check
Riverside Restaurant · 19:30 · 2 covers · available
12:02:01
guest.confirmation.received
Guest approved table on WhatsApp
12:02:02
restaurant.booking.create
Table 2 · 19:30 · confirmation sent
12:04:01
inbound.whatsapp.received
Intent: late checkout · Upsell skill
12:04:02
skills.upsell.evaluate
Late checkout available · €35 · policy OK
12:05:01
guest.confirmation.received
Guest approved late checkout on WhatsApp
12:05:02
reservation.modify
Checkout extended to 14:00 · PMS updated
12:05:02
folio.charge.add
€35 late checkout fee · posted to room automatically
12:05:03
outbound.whatsapp.sent
Late checkout confirmation · audit logged
- Volume
- 100K+ msgs/mo
- Guest match
- Auth + reservation per thread
- Human review
- Per channel toggle
HOW IT WORKS
Four layers. One platform.
From guest intent to system action, the same execution layer runs across messaging, voice, email, PMS, housekeeping, CRM, and audit.
Live execution path
LAYER 01
Skills Engine
Codified hospitality domain logic — reservation matching, VIP escalation, brand voice, and hundreds of operational edge cases.
Booking.com _1 suffix matched · housekeeping ticket created · audit logged
LAYER 02
Agent Runtime
The orchestration core — reasoning, routing, policy enforcement, and execution with full audit trails on every guest action.
Intent classified · rules applied · integration called · outcome logged
LAYER 03
Channel Agents
Messaging, voice, and email — each configured with guardrails, escalation policy, and human-in-the-loop where it matters.
WhatsApp intent parsed · PMS fetch · action executed · guest reply sent
LAYER 04
Integration Fabric
25+ live integrations across PMS, housekeeping, CRM, OTA, and messaging. REST and webhook-driven with per-integration observability.
reservation.find() · ticket.create() · CRM update · channel reply
WHY GENERIC AI FAILS
Hotels run on idiosyncrasies. D3x knows them.
When a Booking.com reservation enters your PMS through the channel manager, the reservation ID gets an _1 appended. A generic AI workflow fails to match the booking. D3x knows.
Skill definition
> reservation_id: BK-48291_1> pms_match: SUCCESS> action: housekeeping.ticket.create
Generic AI
Stops at the answer.
Output
Replies to guest
- Cannot match reservation
- Cannot create ticket
- Cannot escalate VIP
D3x agents
Answers, then executes.
Output
Replies to guest
- Matches reservation
- Creates housekeeping ticket
- Escalates VIP per policy
- Logs to audit trail
Domain depth · 15 years · Codified · Compounds
CONTROLS
Autonomous where it should be. Human-led where it matters.
Late checkout until 2pm for €35?
Awaiting staff review
CONTROL 01
Human-in-the-loop
High-stakes channels operate as co-pilot. Staff review before send.
Riverside Hotel
Response warmth
CONTROL 02
Brand voice configuration
Per-property tone, vocabulary, and response style.
- WhatsAppLive
- VoicePhase 2
- EmailPhase 2
CONTROL 03
Phased deployment
Channel-by-channel rollout. Property-type guardrails.
- 12:04:01guest.authenticate
- 12:04:02reservation.match
- 12:05:02folio.charge.add
CONTROL 04
Audit logs
Full conversation history with decision rationale, queryable for review.
GDPR Compliant (today)·SOC 2 Type II (in progress, Q3 2026)·EU-hosted infrastructure·Data not used to train third-party models
INTEGRATIONS
Live in your stack.
25+ integrations across PMS, housekeeping, channel managers, OTA, table management, CRM, and messaging. Average integration build from API access: 3 weeks.
PMS
HOUSEKEEPING
MESSAGING & CHANNELS
CRM & GUEST EXPERIENCE
HELPDESK
OPERATIONS
IN PRODUCTION
Already shipping. Already scaling.
European hotel groups in production today — aparthotels, hotels, and hostels.
4.58
AI CSAT vs human benchmark
120x
faster median response (15m → 7s)
60%
containment · zero human touch
Staycity Group
75% web chat & WhatsApp handled by AI
“75% of our web chat and WhatsApp messaging is now handled by AI, with high CSAT scores, which are paramount for our business model. D3x is unique in how tightly they can integrate with our systems and workflows for truly seamless 'real world' AI adoption.”
McDreams Hotels
Budget design hotels · Germany
112x faster median response (15m → 8s)
“Operational AI that connects to our existing stack.”
Amano Group
Lifestyle boutique · Germany & UK
5 cities · Germany & the UK
“Agents that execute, not just reply.”
AI LOBBY TALK
AI in European hospitality, from the leaders deploying it.
AI Lobby Talk ✨ — leadership lessons from the front lines of AI in hospitality. Recorded in person with European hotel CIOs and CTOs.


Mike Rawson · CIO · citizenM
Hospitality Is Now a Technology Business

FAQ
Questions CIOs ask before they book.
TALK TO US
30 minutes with the founder.
Not a sales pitch. A working conversation about what AI orchestration looks like at your group's scale, what's deployable in the next quarter, and what we're not ready to claim yet.
Jason · Founder · 15 years operating and building hospitality software.

