Behind the algorithm, how Yango Play’s AI is transforming the GCC digital scene

Behind the algorithm, how Yango Play’s AI is transforming the GCC digital scene

Late last month the tech platform Yango announced the launch of its artificial intelligence (AI) powered entertainment super app – Yango Play in the GCC market. The all-in-one entertainment super app combines – video, games, and music streaming onto one platform. The idea is to provide the users with one comprehensive entertainment experience.  

But how does the platform’s technology engine work, and how does it stand apart from others? One is that it allows users to add up to three other people to one subscription, and connect over 20 devices. While Netflix has been clamping down on shared accounts, Yango Play is opening the landscape for its users.  

On the technology, Roman Shimansky, MENA Region Business Director at Yango Play, said in a conversation with edge/ that the tech platform Yango already has one of the best recommendation algorithms in the region.  

Tailoring the algorithm

The challenge, he added, is tailoring it for a new market, for which the algorithm had no data. “The algorithm not only had to learn quickly, it had to do so on the go,” added Shimansky. He explained the Yango play recommendation engine uses a hybrid approach.  

The engine combines collaborative filtering with content-based models. The collaborative filtering algorithm analyses user actions. “We can find tracks that users who are simliar like and can recommend them. But if we’re in a new country where we don’t have a lot of users yet, we can improve our recommendations using content models. There’s a neural network that analyses the sound itself, it understands whether it’s a song or an instrumental piece,” explained Shimansky.  

Yango has developed its own proprietary GPT technology, large language models (LLMs) and other AI implementations to help retain an edge in the market. “Our extensive experience with AI models spans across various verticals of our product, enhancing everything from our recommendation algorithms to our personalised music feeds and the intuitive voice assistant,” added Shimansky.  

Building differentiated AI models

These advanced AI technologies refine the user experience and stay ahead in the rapidly changing entertainment domain. The AI-driven recommendation systems are not solely reliant on algorithmic predictions but also take in industry insights gained from Yango’s presence in the entertainment sector.  

Shimansky said the unique blend allows them to fine-tune the technologies to meet and anticipate the evolving tastes and preferences of the users. “Our strength lies in recognising which technology and product format would provide the most value to users and the overall market,” he added.  

While there is a strong tech development culture within Yango, the team understands that developing deep tech from scratch may not necessarily be efficient every time. The team therefore identifies the best benchmark technologies and creates the best user experience with them.  

“In the case of Yango Play, we licensed the best recommendation technologies from our partner, Yandex, and created something entirely new to any region — the first AI-powered entertainment service that combines video streaming, music, and mini-games in one place,” explained Shimansky.  

The team ensured enough market research and conversations with the potential audience to help develop a modern entertainment service. “As technologies evolve, people’s idea of a good product evolves too. This meant spending months crystallising the vision of product concepts and experimenting with different conceptual versions of the app, which resulted in Yango Play.  

Understanding user models

During the prototype stage, the team also checked whether the users liked their vision and whether the product was intuitive and easy to use. “After that, we put all that groundwork together to bring the product to life and create something completely unheard of: an all-in-one entertainment app where you can watch movies and series, listen to music, play games, and more,” adds Shimansky.  

He explained the core strength lies in the ability to identify the best technologies and create superior services. Added to this is a deeper understanding of region-specific needs.  

“We focus on the local communities’ key requirements for products and services, and we tailor our benchmark technologies to those specific requirements,” added Shimansky.  

Currently, the team believes Machine Learning and AI will continue to be an important factor, and will be a significant part of the creator landscape. “Other technologies that have been gaining momentum are AR and VR. Those are more about video entertainment, but we’ll surely be able to create immersive user experiences, more so than ever before,” added Shimansky.  

Speaking of the plans, Shimansky said, “Our primary goal is to develop more services that millions of people will use daily across the region, making their lives a bit easier and more joyful.”