Global beauty product manufacturer Coty asked for our support to demonstrate a recommendation engine in Luton airport’s duty-free zone.

The client asked for more than one customer to be able to use the app at the same time. They also wished to capture the resulting data so the recommendation engine could learn and train itself throughout the demo.

We were not able to know the minimum or maximum number of customers that would be using the software simultaneously. This presented a risk of either overspending on unnecessary IT resources if there were fewer users than planned, or offering a poor user experience otherwise.


We decided to use Microsoft Azure cloud computing, which we could scale up and down from as and when needed. This way, user experience is the same whether two or 20 people are running the software engine at once. This also meant we would only pay for the IT resources we actually used.

To create an app that would be technically sound, scalable and able to collect the data needed, we combined Azure with Docker, a computer program providing ‘container’ cloud-enabling technology.

For the demo, we used a test base of four iPads.


We were able to operate our data science models in a live environment and collect data through the recommendation engine

We successfully deployed the very latest technology – containing highly intensive computational algorithms – to an application

We ran it efficiently on multiple devices at the same time

More client stories

Connect with us

Fire us a message and we’ll be in touch

Contact us