Find the right AI use cases before investing in development.
Connect AI to your existing systems and workflows.
Build autonomous AI agents that automate complex tasks.
Design and develop AI solutions tailored to your business.
Build secure, scalable AI systems for enterprise environments.
Legacy software modernization →
Upgrade outdated systems for better performance, security, and scalability.
Make informed technology decisions before investing in development.
Validate your product idea with user and business insights.
Define the right scope to launch faster and reduce risk.
Confirm your idea is technically viable before you build.
Capabilities
Want to start a software project with us?
Book a time with our technology experts →
We brought young developers together to explore the future of AI agents. More editions are coming soon.
An AI concierge app for a European hotel group, built with agentic workflows and shipped as an MVP in four months. 2,500 downloads in month one, from in-room QR codes alone.
Total installs in the first month
Average session length
Daily app opens
App store rating
Hospitality and travel
EU
Product discovery
Mobile app development
AI integration
Technical Project Manager
7 Software Engineers
3 QA Engineers
4 months
Our client is a European hospitality group running hotels, resorts, and camps. Their guest data, content, and operations was in several disconnected systems, with nothing that reached guests directly. They wanted one app to pull it together: content, loyalty, and an AI concierge that truly improves the guest experience. We joined as their engineering partner.
Building a genuinely useful AI-powered guest app meant clearing a few real obstacles:
We built a guest app with an AI concierge at its core that gives guests one place to plan their stay, ask questions, and spend on property:
We set three primary goals: more app downloads, more in-hotel micro-transactions, and a better guest experience. Every later decision had a clear reference point.
We audited the group’s property-management, CRM, operations, and knowledge-base systems to map exactly what the gateway would aggregate and what we would store versus pass through.
We expanded our presales estimate into a defined scope, then decided what belonged in the first release and what could wait. The build was mapped into milestones, each a dated set of features, so the scope doubled as a delivery schedule.
We ran the build against a project harness of rule files, standards, a self-review agent, and an onboarding agent. New team members became productive almost immediately, and the full backlog was generated through an AI and Jira workflow.
A thin aggregation layer in Node.js pulls guest and property data from the client’s existing systems and serves it to the app. We store very little ourselves, acting as connective tissue across the group’s stack.
A Payload CMS admin panel lets our client manage all guest-facing content, served to the app per property and per page.
A vector database in PostgreSQL/pgvector, synced with our client’s central knowledge base, powers the concierge. We built an evaluation and prompt-testing loop to keep answers accurate.
Guests can browse the client’s properties and offers by region, by brand, and by the kind of trip they’re after, from an active vacation to a family one. It’s how a guest finds their next stay, or gets to know the one they’ve already booked.
A guest enters the booking code from a reservation they already hold, and the app connects to it. Content narrows to their property and region and the concierge scopes its answers to that hotel. Unlinked, the app still works as a catalog of their properties.
The concierge answers from our client’s knowledge base, the same source their call center works from, so a guest asking when breakfast runs or what their rate covers gets real answers. We built an evaluation loop that scores its answers against test cases, so it stays accurate.
Driving in-stay spend was the point of the app. Each day it shows fresh offers with images and a notification to match. Guests see what’s on without hunting for a menu, and the client gets a direct channel to in-stay revenue.
Guests can check what they’ve spent and bought at any point in their stay, without a trip to the desk.
Members collect points and unlock perks and in-app discounts they won’t get elsewhere. It gives our client a direct line to every guest: a reason to open the app, and a reason to come back.
Our client came to us with several disconnected systems and no way to reach guests directly. They now run a single app, built on one gateway that ties those systems together, with the groundwork to modernize their other systems:
This is the first project where AI wrote the code, inside agentic workflows we defined. Senior engineers owned every feature, and everything passed a review agent and our QA team before shipping. The result? Ten months of work done in four.
Most hotel and resort groups have guest data in one system, content in another, and no single channel to reach guests during their stay. If you’re ready to bring it together, fill in the short form below or book a call directly with our client strategy specialist.
Don’t worry, any information you give us will only be stored for business purposes. Find out more in our Privacy Policy.
Arrange a chat with Ivor, our client strategy specialist.
Communication and collaboration app for 70.000 crew members across 40 ships.
Backend booking engine that extends Salesforce with real-time pricing and configurable travel packages.
