Restaurant Finder application for NYC demonstrating Atlas Search Features. Uses React and Tailwind.

What's Cooking implements many Atlas Search features from autocomplete to custom function scoring. Using the $search operator in a MongoDB aggregation pipeline, we can build fine-grained searches across text, numerics, and geospatial data. By building out What's Cooking, you'll learn all sorts of ways MongoDB allows you to build complex, fine-grained full-text searches on your Atlas data.


No additional servers or software needed. No need to keep data in sync. Everything is done in MongoDB Atlas.

  • fuzzy matching
  • highlighting
  • autocomplete
  • range queries
  • geoqueries
  • facets
  • relevance-based scoring
  • custom function scoring

Check out the video of the MongoDB .Live keynote to see a demonstration of all the features or visit the link below to play around with the finished application, hosted entirely in MongoDB Atlas:


This application is hosted entirely by MongoDB Atlas was created using:

  • React
  • Tailwind CSS
  • A modified sample dataset based on MongoDB's Atlas sample_restaurants dataset

Currently this app is not suitable for mobile, but feel free to send a PR. ?


  • A MongoDB Atlas account. Get one for free here.
  • A recent version of Node.js and npm.
  • Restaurant sample dataset.
  • (Recommended) MongoDB Compass - GUI

You can read and download the dataset using the MongoDB Shell, any MongoDB driver, or my favorite MongoDB Compass using the following URI:

mongodb+srv://mongodb:[email protected]/whatscooking

It is also included in this repo's Supplemental Files branch as


To Run This Application....

  1. Clone the repo.
  2. Navigate inside WhatsCooking directory.
  3. Run npm install .
  4. Run npm start .

To Build This Application...

Prepare Data

  1. Load data to Atlas cluster:
    • database: whatscooking
    • collection: restaurants
  2. Create Search indexes. (Index definitions includes in `supplement-files` .)

React Components....

Using Realm as Your Serverless Backend....

What's Cooking uses HTTP services in Realm to create 2 APIs to allow you to query for your restaurant data over HTTP:

  • getRestaurants called from the useHomeFetch.js hook.
  • getRestaurantsAutocomplete called from the Autosuggestions.js component.

Find the Realm application and code for these webhooks in the supplement-files branch in a folder called WhatsCookingRealm

If you have any questions or feedback about this repo, feel free to create an Issue or PR in this repo or reach out to me on Twitter @YouOldMaid.