Meesho open-sources its BharatMLStack to boost AI innovation

The social ecommerce platform Meesho has made the key components of its internal machine learning platform, BharatMLStack, available as open-source to enhance the accessibility of AI infrastructure for developers. This release encompasses its online feature store, control plane, orchestration UI, and SDKs – all of which are now accessible on GitHub.

The IPO-bound firm reported that its BharatMLStack, which has been developed over the past two years, facilitated over 66.9 trillion feature retrievals (data signals obtained for real-time predictions) and 3.12 trillion real-time inferences (real-time predictions) at its peak during FY25. The ML architecture, which handles Meesho’s platform used by over 2 million sellers and 160 million-plus consumers, was open-sourced. “We believe great technology should scale impact, not just infrastructure,” said Sanjeev Kumar, Founder & CTO of Meesho. 

“We put BharatMLStack to the test during high-traffic events like our Mega Blockbuster Sale in March 2025, where it delivered at scale, demonstrating its ability to perform under peak load conditions. This helped Meesho drive higher user engagement, better conversions, and increased order volumes during the sale. By open-sourcing it, we’re sharing a high-scale, AI stack with the broader tech community—purpose-built for real-time use cases and tailored for Indian businesses,” he added.

The new platform features an Online Feature Store that helps AI systems by serving pre-computed, up-to-date features instantly for real-time model predictions, enabling use cases like fraud detection and personalised recommendations to run faster and more accurately. By sharing this with the broader tech ecosystem, Meesho aims to empower startups, ML engineers, and data scientists across India with access to production-ready, cost-efficient AI tooling tailored for India’s fast-evolving digital economy. 

According to Kumar, teams can collaborate more effectively, avoid data silos, and quickly build or enhance AI models with consistent, high-quality features, all while accelerating deployment and driving innovation. While it doesn’t replace model development, it significantly simplifies training, deployment and infrastructure management, allowing data scientists and engineers to focus on what matters most: building better models and accelerating innovation.

The company said future components of the stack will also be open-sourced, with opportunities for the developer community to contribute and iterate together. 

Meesho is not looking to monetise the release at this stage. Instead, the move is part of a broader attempt to support India’s growing ML developer base and to gather feedback and contributions that could improve the stack further.

In recent months, Zerodha unveiled a $1 million fund dedicated to investing in open-source projects. Zomato has recently made its weather monitoring system and design system, Sushi, available as open source. In a similar vein, A public repository on Flipkart's GitHub, which has garnered over 5,000 stars, is named RecyclerList View and offers a high-performance list view component for both React Native and web applications.