Build Real Things
Plug into Swiggy’s live APIs - Food, Instamart, Dineout - and build things people actually use. Real data, real users.
For Developers
Real APIs. Real users. Build AI agents and wild experiments on Swiggy's MCP platform.
No approval needed to start. Build locally for free, then apply only when you're ready to go live in production.
Why Developers Love This
AI-Native Platform
MCP is designed for agents. Build AI assistants, copilots, and automations that order food, manage groceries, or book tables for users.
Hack & Experiment
Move fast, test wild ideas, and iterate quickly. We give you the space and the tools to try bold things.
Get Hired by Swiggy
Build something impressive and send us a demo. Standout projects get featured - and yes, we actively hire from this program.
What Could You Build?
Starting points, not limits. These are the kinds of agents and tools that work well on Swiggy's MCP.
01
Voice Agent
Conversational AI that handles end-to-end food ordering via natural language
See the voice pattern02Auto-Restock
Instamart agent that learns household consumption patterns and reorders before you run out
Grocery recipe03Group Ordering
Slack/Teams bot that collects lunch preferences from a team and places a single optimized order
Food recipe04Dietary Planner
Meal planning copilot that filters menus by macros, allergies, and calorie targets
search_menu reference05Reservation Agent
Dineout integration that matches group availability, cuisine preferences, and budget constraints
Book-a-table recipe06Multi-Modal Agent
Screenshot-to-order pipeline: snap a dish photo, identify it, find it on Swiggy, add to cart
Framework recipesYour Toolkit
Three MCP servers packed with tools. Pick one or use all three.
0120 tools
Search restaurants, browse menus, manage carts, and place food delivery orders. Build AI agents that handle the entire ordering flow.
- search_restaurants
- search_menu
- get_restaurant_menu
- update_food_cart
- get_food_cart
- place_food_order
- track_food_order
0219 tools
Search grocery products, manage cart, checkout, and track instant deliveries. Perfect for building smart shopping agents and auto-reorder bots.
- search_products
- update_cart
- get_cart
- checkout
- track_order
- get_orders
0312 tools
Search dine-in restaurants, check availability, and book tables. Build dining recommendation agents and smart reservation tools.
- search_restaurants_dineout
- get_restaurant_details
- get_available_slots
- book_table
- get_booking_status
We actually hire from this program.
Ship a great project, send us a demo, and our recruiting team might reach out. No whiteboard puzzles. No trick questions. Just show us what you can build.
Send Us a Demo- 01
You build something cool on Swiggy’s APIs
- 02
You send us a demo or a GitHub link
- 03
We feature it across our channels
- 04
Our recruiting team reaches out
Developer FAQ
The stuff you're probably wondering.
No. You can start immediately - prototype on localhost with no approval and no cost. You only apply for access when you’re ready to take your integration live in production.
Yes. Start building right away - no approval needed. We welcome individual developers, indie hackers, and small teams; when you’re ready for production, apply and tell us about your use case. Some of our most interesting projects have come from solo developers.
You get access to all three: Swiggy Food (restaurant search, menu browsing, cart management, order placement, tracking), Instamart (grocery search, cart, checkout, order tracking), and Dineout (restaurant search, slot availability, table booking). Each server exposes a set of tools you can compose into agent workflows.
After approval, you receive API credentials and redirect URIs for your integration. Authentication flows are handled via the MCP protocol - your agent authenticates on behalf of the user, and all API calls are scoped to that session. We provide documentation and examples for common auth patterns.
Absolutely. That’s where it gets interesting. An agent could search Dineout for a restaurant, check Instamart for ingredients to cook the same cuisine at home, and offer both options to the user. Cross-server workflows are fully supported and encouraged.
Show us the full loop: a real use case, the agent workflow, and a working demo (video, deployed app, or GitHub repo with clear setup instructions). We care about product thinking as much as technical execution. Email demos to builders@swiggy.in.
You work directly with production APIs - real restaurants, real menus, real inventory. This is intentional: we want you building with real data so your agents handle real-world edge cases from day one. Rate limits and order caps keep things safe during development.
05 / BUILD TOGETHER
What Will You Cook?
Get API access, build with real Swiggy data, and show us what you've got.