Meesho's AI System PRISM Drives Intent-Driven Discovery for Millions of Indian Shoppers

Meesho's AI System PRISM Drives Intent-Driven Discovery for Millions of Indian Shoppers​

Bengaluru: Meesho, a leading e-commerce platform in India by Annual Transacting Users and Placed Orders, has highlighted the significant role of its proprietary artificial intelligence system, PRISM (Personalised Ranking & Intent Signal Module). This technology is designed to fundamentally change how millions of Indians discover products and shop online on the platform.

PRISM addresses the reality that for large segments of the Indian consumer base, shopping resembles browsing a local bazaar rather than executing a fixed keyword search. For Meesho, which serves 264 million annual transacting users and recorded 717 million placed orders in Q4 FY26, this discovery-led approach is critical. Currently, over 75% of all orders on Meesho originate from AI-driven personalized feeds powered by PRISM.

The platform's design shifts commerce away from traditional keyword-based search toward real-time intent understanding and product discovery.

Technical Architecture and Capabilities​

PRISM operates on a sophisticated real-time intelligence and ranking architecture built specifically to understand the nuances of online shopping across India ("Bharat"). The system continuously interprets behavioural, transactional, and contextual signals generated by users throughout their browsing journeys.

Key operational specifications of PRISM include:


The system is supported by BharatMLStack, Meesho's in-house ML infrastructure platform. This technology enables high-throughput AI workloads while maintaining significantly lower inference costs compared to conventional cloud infrastructure. Furthermore, PRISM supports over 10 multilingual experiences, including Hindi, Bengali, Marathi, Tamil, Telugu, Kannada, Malayalam, Gujarati, Punjabi, and Odia.

Through Trendpulse, an LLM-powered discovery engine integrated into PRISM, the platform interprets emerging demand patterns across various regions, cities, and local consumer clusters, ensuring that highly relevant products are surfaced aligned with India's diverse evolving shopping behaviours.

The Shift to Discovery​

Debdoot Mukherjee, Chief Data Scientist, Head of AI and Demand Engineering at Meesho, commented on the transformative nature of this shift. "The next hundred million Indians coming online will not search, they will discover. They will not type, they will speak, browse, and expect technology to meet them where they are," he stated.

Mukherjee added that PRISM integrates real-time intelligence and ranking systems to help consumers express intent and make purchase decisions online. The system is designed for "consumers who are discovery-first, multilingual, and increasingly voice-led," pre-empting the need for a consumer to articulate exactly what they want.

PRISM already functions across multiple areas of Meesho, powering personalized recommendations, trend intelligence, catalogue understanding, intent-led discovery, and seller growth systems. This functionality is amplified by the scale of the Meesho ecosystem, which includes 263 million monthly active users, 17 billion product views per day, 1,592 million ratings, and 505 million reviews.

As discovery continues to grow across India, PRISM aims to power future shopping experiences on Meesho, enabling sellers to reach demand efficiently and continuously compounding intelligence across billions of real-world interactions.

MEESHO Stock Price Movement​

Shares of Meesho Limited today slipped by 6.00% to settle at ₹165.34 after trading closed. The equity saw considerable activity as 14.19 million shares were traded during the session.