With AWS technology, iFood implements AI area to enhance client and restaurants experience

Over a year ago iFood, leading player in the Latin American food delivery market, decided that it was time to invest in Artificial Intelligence in order to be more assertive in restaurants and consumers service. To support the new structure, as well as its demand for processing high data volumes, the company relies on the services provided by Amazon Web Services, which offer the flexibility and scalability required to provide information in real time on a number of the company´s operations.

Founded in 2011, iFood, a foodtech leader in Latin America, operates in Mexico and Colombia and, in its beginning years, improved its operating model, following the evolution in the market and growing bigger year by year.

Initially, iFood operated as an integrated network for fulfilling online orders from the restaurants that took part in it. As time went on, the company started developing its own technology. Eighteen months ago, for example, they came to work with its own fleet, through deliverymen that register with the platform, and which today account for a significant part of the deliveries.

Nowadays, more than 20% of the orders placed use the entire iFood´s sales platform, from their own marketplace to CRM to the deliveryman. “There were already AI initiatives in the company, but they were not centralized, and were used basically to answer to questions on business areas”, recalls Sandor Caetano, Chief Data Scientist at iFood, pointing out that the company already relied on a data structure.

Over a year ago, iFood decided to create their Artificial Intelligence Academy, focusing on the development of research on machine learning, deep learning, logistical efficiency and other areas related to the company´s ecosystem. The initiative is the result of an investment of 20 million USD, as part of an intake made by shareholders in 2019.

“We created this area to meet the need for growing, escalating and facing the competition”, explains Caetano, recalling that the goal was to count on a structure that would take into consideration the existing AI team and started to support the decisions of the company board. “That would change the way machine learning was dealt with, since it would start to be used for the automation of decisions, providing plain answers where an excess of data was in place”, he said.

In order to perform the developments, make the necessary tests and put them to work, iFood would need a robust TI infrastructure that, nonetheless, would be fully utilized in peak moments only. The option was to search for a company that would offer such infrastructure in the cloud, enabling a pay per use model and providing flexibility and scalability in case of need.

iFood already used AWS services like storage and managed database, among others. Thus, expanding the work scope was the natural path. “The structure offered by AWS allows us to train several models in machines sized the way we need and, more than that, to put these models into production in a simple way, without needing to start from scratch. Moreover, we rely on tools that make the data scientist´s life much easier, making him the most productive possible”, says Caetano.

These are some of the main solutions used: Amazon SageMaker, which holds all the model testing infrastructure; Amazon Kinesis, which makes streaming data collection, processing and analysis easier in real time; and Amazon Elastic Kubernetes Service (Amazon EKS) to keep the models on air.

The use of AWS infrastructure enabled iFood´s Artificial Intelligence Academy to expand the use of algorithms. One of the first areas to benefit from it was Logistics. “We need to tell the client how long it will take for his meal to be delivered”, says he. To this end, iFood relies currently on a route simulator, where it is possible to analyze different operation parameters according to week days and times.

Simulations are carried out in AWS safe environment and, once proved, they are put into production. “The use of such algorithms sped up the process of discovering new parameters and, thus, our Logistics area is far more efficient today. We deliver more today than two years ago”, compares Caetano.

The same applies for the recommendation lists to clients. iFood app recommends restaurants and dishes according to user´s taste by means of models processed at AWS. “All this goes through these models. Today we are able to control what is going to be shown in the app, strategies and promotions, thanks to these algorithms. Such level of service customization is possible thanks to Amazon SageMaker, which processes all of our models”, he emphasizes.

Today, all of our models tested by iFood´s Artificial Intelligence Academy get on the air with a benchmark already in place. Hence, the company has been obtaining expressive gains in terms of productivity and improvements in its service levels. According to Caetano, since the algorithms started to be used, the delivery SLA increased from 80% to 95%.

In the logistics area, the distance traveled by the deliverymen had a 12% reduction thanks to the optimization of the routes. The same happened to the idle time of these professionals, which was reduced by 50%.

“When it comes to recommendations, there was also a significant improvement in the conversion”, celebrates Caetano, in that today it is possible for iFood to optimize lists so that they appoint restaurants near to clients houses, thus optimizing the deliveries as well.

Besides, the company obtained gains such as:

• The company reached more than a thousand cities;
• The ecosystem reached 220 thousand restaurants and 170 thousand registered deliverymen using the platform;
• iFood completes over 39 million orders per month.

With over a year of operation, the Academy should continue to expand its activities. According to Caetano, focus is on the horizontal expansion. “We are still in the beginning. There are applications which have been developed by us in order to make restaurants life easier”, reveals him, having in mind that, here, there are two different approaches. One of them provides for the use of AI in order to improve the image of the dishes that the restaurants display in the application. In another one, Artificial Intelligence will be used to aggregate data to the description of such images, enabling the client to identify the ingredients of the dishes, for example.

Get more information on AWS artificial intelligence solutions.