In Swiggy's Developer Community, Some are Planning Parties, Other are Prepping Meals

For years, food delivery apps have largely worked the same way. Users open an app, browse hundreds of restaurants, compare dishes, add items to a cart, and finally place an order.

Artificial intelligence could soon change that. Instead of manually searching for meals, users may simply tell an AI assistant: “Order me a high-protein dinner under 600 calories,” or “Get groceries for my week’s meal prep.”

Swiggy believes this is where commerce is headed. In April this year, it launched Builders Club, opening its commerce infrastructure to developers and allowing them to build AI agents that can recommend meals, plan nutrition, enable voice ordering for elderly and visually impaired users, and complete transactions on a user’s behalf.

Swiggy Builders Club enables access to three MCP servers and more than 18 API tools. Built on Amazon Web Services, the programme is powered by Amazon Bedrock, which offers unified API access to foundation models from Anthropic, Meta, Mistral AI, and others, alongside AWS Trainium.

From Food Delivery to Food Decisions

One of the biggest trends emerging from Builders Club has been the rise of health and nutrition assistants. Developers are building AI agents that understand dietary goals and make recommendations accordingly.

“One thing we’ve learned since opening Builders Club is that developers often see opportunities we wouldn’t have thought of ourselves,” Vivek Garg, Vice President of Engineering at Swiggy, tells AIM.

“Health and nutrition have quickly become one of the most active areas. We’re seeing meal planners built around calorie or protein targets, agents that can turn a nutrition plan into a grocery cart, and assistants designed for people managing specific health conditions,” he adds.

Unlike traditional recommendation systems, these assistants understand users' intent before they search for food.

A user may ask for a high-protein dinner after a workout, groceries for a keto diet, or ingredients for healthy home cooking. 

“The agent understands that intent, finds the right products through Swiggy, builds the cart and completes the transaction. We handle catalogues, authentication, payments, and fulfilment allowing developers to focus on solving the user’s problem instead of rebuilding commerce infrastructure,” he explains.

Making Food Ordering More Accessible

Another area gaining momentum is accessibility. Developers are building voice-first AI assistants for users, particularly elderly users and people with visual impairments, who find mobile apps difficult to navigate.

Instead of scrolling through menus or typing searches, users can simply speak naturally.

“Some of the most meaningful projects have come from accessibility,” says Garg. “Even WhatsApp-based personal assistants that help people order food or groceries without navigating an app.”

The AI agent handles everything behind the scenes. A user might simply say, “Order my usual,” or “Get this week’s groceries.” The assistant identifies previous preferences, builds the order, and places it through Swiggy.

“Instead of learning how to use an app, people can simply express what they need. For someone who is visually impaired, elderly or not comfortable with digital interfaces, removing that friction makes technology much more accessible,” he explains.

For Swiggy, these accessibility projects represent something larger than convenience. “AI quietly removes a barrier. Technology fades into the background and simply helps people get things done,” he persists. 

Building AI for Everyday Life

Builders Club is already producing experiments beyond food delivery.

One such project is tanhAI, built by Rachit Yadav, an associate AI product manager at a tech company. His AI assistant acts as a social concierge, helping users plan parties, recommend cocktails, suggest restaurants, and order missing ingredients through conversation.

“The idea came from a house party. A friend wanted to make a cocktail he’d seen on Instagram months ago, but nobody remembered the recipe. That’s when I realised there was a gap between discovering something interesting and actually acting on it,” says Yadav.

Using Swiggy’s Food, Instamart, and Dineout MCP servers, tanhAI can recommend recipes, identify missing ingredients, place grocery orders, suggest restaurants, and even coordinate an evening out all through a single conversation.

“The interesting part is that users don’t think about switching between services. They’re just having a conversation while different Swiggy services work together behind the scenes,” he adds.

He notes that Builders Club allowed him to focus on building the user experience instead of spending months creating commerce integrations from scratch.

“The documentation has been genuinely good, and the Swiggy team has been extremely approachable. I could have built it from scratch, but it would’ve taken much longer, and I probably wouldn’t have been able to iterate this quickly,” he asserts.

Swiggy’s Bigger Bet

For both Swiggy and developers, the bigger opportunity lies in conversational commerce. Instead of opening multiple apps, users may increasingly rely on AI assistants to understand their intent and complete tasks automatically.

However, traditional apps may not disappear altogether.

“I see apps evolving into execution and discovery layers, while conversational agents become the interface people use for convenience,” Yadav notes.

In other words, people may still browse restaurants when they want to explore new places, but routine decisions could increasingly be delegated to AI.

Since announcing Builders Club, Swiggy has received over 2,000 developer applications and has started onboarding its first cohort, gaining access to the platform’s vast commerce network.

Rather than building payment systems, product catalogues, and delivery infrastructure from scratch, it wants to enable developers to focus entirely on creating new AI experiences.

For Swiggy, that’s the larger vision. “The company is no longer thinking only about food delivery. It is preparing for a future where AI agents become the primary interface through which people interact with commerce,” opines Garg.