By Kavita Mistry 
From being at the helm of “pulse of Bengaluru”, to being the at the forefront of science, Deepinder Goyal is treading a path where technology meets tomorrow—investing in life, engines, and ideas that dare to reshape the future.
From Food to the Future
Recently, Deepinder Goyal has signalled his focus on building tech-driven, future-facing ventures. His $25 million personal commitment to Continue, a research initiative supporting global studies on human ageing, highlights his investment in health science.
On the other hand, he launched LAT Aerospace in Bengaluru to develop indigenous gas-turbine engines for India.
Amid all this, Goyal is not forgetting his first and biggest taste of entrepreneurial success, “Zomato” and is extending his vision to Zomato as well. AI is currently playing a central role in Zomato.
Zomato is consistently exploring AI applications, be it – Nugget – an AI-powered customer support platform handling millions of interactions, or for that matter – Espresso – an open-source PDF-generation and signing tool.
Last week, the company quietly released an official MCP server on GitHub, leveraging the open-source framework from Anthropic. MCP connects AI systems and chatbots to live data, allowing them not just to read but also interact with external services through approved interfaces.
For the uninitiated, MCP acts as a bridge between AI models and live data or services.
How Does It Work for Zomato?
For Zomato, this enables users to perform tasks via natural language that would normally require manual input: searching restaurants, browsing menus, adding items to carts, ordering, and completing payments through QR codes. MCP access is available via ChatGPT or Claude, though it requires manual setup.
Shubham Palriwala of Agnost AI highlighted how a single prompt like “I’m hungry, get me something spicy under 300 rupees near MG Road” can fetch options, ratings, and delivery times, reducing multiple clicks into a single conversation. Lastly, payment is finalised via QR scanning to ensure security.
There are other Indian startups that includes the likes of Zerodha and Razorpay that have also integrated MCP, for connecting to financial portfolios and handling payment actions respectively.
Experts see MCP as a potential revival of the “super app” model, simplifying complex multi-step workflows. However, developers must balance dynamic tool access with reliable execution to prevent AI errors or hallucinations. MCP’s strength lies not just in fetching data but in taking contextual actions, unlike frameworks such as RAG or GraphQL.
Globally, organizations including OpenAI, IBM, Google, and Microsoft have launched MCP solutions, while developers are building tools for autonomous actions in AI chatbots. Anthropic and OpenAI’s Apps SDK further allow third-party services to integrate directly into chat interfaces, making AI interactions more seamless and functional.
Summary
Deepinder Goyal is steering his vision toward future-focused, tech-driven ventures—investing $25 million in Continue for longevity research and launching LAT Aerospace for indigenous engine development. On the other hand, his crown jewel, Zomato advances in AI with tools like Nugget, Espresso, and a new MCP server enabling natural-language restaurant searches, ordering, and payments—showcasing Goyal’s pursuit of innovation where technology meets tomorrow.