JioStar's Chief Architect on Scaling AI for 500M Users
JioStar, a leading entertainment and live-sports platform with over 500 million monthly users, has redefined content discovery by integrating AI-powered conversational interfaces. Vijay Seshadri, Chief Architect at JioStar, shared insights into the development and deployment of this system during OpenAI’s Executive Summit in Singapore.
AI-Powered Content Discovery at Scale
JioStar’s conversational assistant, integrated into its JioHotstar platform, allows users to discover movies, shows, live-match moments, and cricket statistics through natural language interaction. The system, built using OpenAI’s API and Codex, supports over 15 languages and was developed in just eight weeks. This innovation addresses the limitations of traditional recommendation systems by enabling users to express nuanced preferences, such as requesting content to match specific moods.
"Content discovery is inherently visual, but combining those visual options with spoken preferences and conversation creates a powerful new paradigm," said Seshadri. The platform’s conversational discovery feature has already seen significant adoption, with over 60% of users opting for voice search over text, according to a report by cmotech.in on June 25, 2026.
Engineering Challenges and Achievements
JioStar’s system has been designed to handle the demands of its massive user base, including peak moments of 72.5 million concurrent live-sports viewers. The engineering team prioritized speed, reliability, and scalability to ensure seamless performance. At peak times, the platform has processed nearly 5 million subscription requests per minute.
In addition to improving user engagement and watch time, the conversational interface has the potential to influence content production by providing real-time audience feedback. Bharath Ram, JioStar’s Chief Product Officer, emphasized the strategic importance of this shift, stating, "The future is conversational search. Interesting search results lead people to watch content they otherwise wouldn’t have watched, increasing engagement, watch time, and our ability to monetize."
Lessons for AI Deployment
Seshadri highlighted the importance of effective evaluation systems in production, noting that balancing trade-offs such as speed and accuracy was critical at JioStar’s scale. He also stressed the value of small, focused teams in achieving rapid development timelines. Reflecting on the project, Seshadri advised executives to approach AI deployment with a clear focus on scalability and user-centric design.
JioStar’s success demonstrates how AI can transform content discovery, offering a blueprint for other platforms aiming to enhance user experiences and operational efficiency through conversational interfaces.