The data platform that finally shipped.
Marjin AI pulls a hospitality business's scattered data into one place and lets its people ask questions of it in plain language. The founder had a validated idea and buyers waiting to see it. What he didn't have, after several attempts with different development teams, was a product he could put in front of them.
The story
The idea itself had already been validated with the market. Getting it built was the problem: a series of development teams had tried and none had shipped, and the most recent attempt, built entirely in Lovable, left the business with an expensive prototype that didn't work and tried to do a million different things badly.
We took the project on in partnership with Velocity, who led product development and design while we handled the engineering. Together we started with an Intention Audit: a full review of the existing system and every feature in it, measured against what the business was actually trying to achieve. From that came a way forward. We helped the founder narrow the product down to a few features worth building well, and laid out a roadmap that delivered them in order of priority, so a real application could reach potential clients as soon as possible.
Twelve months later the base product was complete, an internal engineering team had been hired and trained, and we handed the platform over to them so the founders could move from product development into sales mode. Below is how we got there.
What we built
Two builds ran in conjunction: a data platform able to ingest hospitality data at national scale, and the full application that turns it into answers. Everything below shipped inside the same year-long engagement, with Velocity leading product development and design throughout.
The engagement started with a full system and feature audit of the existing Lovable prototype, measured against what the founder was actually trying to achieve. The prototype attempted a great many things and did none of them well, so the audit separated the intent worth keeping from the build that was holding it back. Its output was a concrete proposal: what to keep, what to cut, and what to build first.
Together with the founder we cut the sprawl down to a small set of genuinely valuable features, then sequenced them into a year-long roadmap ordered by priority. The aim was to get a working application into potential clients' hands as early as possible, and to let every feature after that earn its place.
We built a full ETL pipeline with multiple routes into the platform: API polling, webhook ingestion and file polling among them. Hospitality data arrives from finance systems, EPOS tills, CRMs and raw file uploads, each in its own shape, and the pipeline was engineered for mass-market operators running hundreds of sites.
Ingested data lands in a bronze layer, is cleaned and normalised through silver, and arrives in a gold layer modelled around the questions the business needs answered. The application and its AI features query that gold layer only, so every insight is drawn from one consistent, trustworthy view of the data.
In parallel with the data work we built the complete application on top of it: financial management and business management features that turn the pipeline into day-to-day answers for operators, from revenue and cost tracking to the operational picture across sites.
The most interesting technical problem in the build. Insights are produced by a multi-agent LLM process engineered to enterprise compliance and quality standards, then optimised over time for accuracy, intelligence and cost together. Questions are translated into database queries so the model reasons over small result sets rather than raw data, and caching means a repeated question never pays for inference twice.
Value delivered
The deliverable was never just software. It was a business that could finally stop building and start selling, with its own people running the platform we left behind.
One Eleven takes your app idea and quickly turns it into a solid plan, delivering top-notch results fast, and keeping you part of the process throughout.
Rob McEwan Co-Founder, Marjin AI The toolkit
We start every relationship with a free consultation: an honest conversation about what you are trying to do. No pitch, no obligation.
Book a free consultation
