Swiggy is expanding its commerce services to external AI agents, allowing users to discover food, build carts, place orders, and book restaurant tables through AI assistants outside its app, the company revealed at Inc42’s CTO Summit 2026. The company has deployed four Model Context Protocol (MCP) servers that connect 66 tools across its food, Instamart, Dineout, and Scenes offerings to enable this integration, according to inc42.com.
Swiggy CTO Madhusudhan Rao explained that the MCP servers act as connectors enabling AI tools to interact with Swiggy’s commerce infrastructure. This setup allows external AI assistants to handle user requests by tapping into Swiggy’s services for discovery, menus, carts, ordering, reservations, and tracking. Rao highlighted that users can link Swiggy’s MCP with any large language model (LLM) to, for example, plan meals around dietary schedules, demonstrating the flexibility of the system, inc42.com reported.
This move reflects a shift from expecting all customer journeys to start within Swiggy’s app to making core commerce capabilities accessible to AI agents. By exposing its services via MCP, Swiggy avoids building separate interfaces for every use case and enables personalized experiences through AI. This approach positions Swiggy alongside other companies leveraging AI to extend service accessibility and customization, enhancing user convenience and operational scalability, inc42.com noted.
Swiggy’s current MCP infrastructure includes four servers and 66 tools spanning various commerce functions. The company continues to focus on building simple, core capabilities that can be invoked by humans, services, or AI agents, as Madhusudhan Rao stated at the CTO Summit, according to inc42.com.