There are many frameworks out there that claim to be the "easiest" or "fastest" to build something. This assessment is designed to test a set of capabilities and their developer experience to objectively determine ease of use.
In this assessment, you will create an API that ingests GPS data and a webpage that displays it. We go beyond a simple CRUD app to veer off the happy path into a real-world scenario that involves events, real-time notifications, and an API.
- Build an API that implements the API specification included in the repo
- When a new session is created, it must store a creation timestamp and an
activestatus. Then return an identifier to the caller - Data points must be added to valid sessions in an
activestatus - Data points must be their own record in the database and not a collection hanging on the session record itself
- When the first data point is added, asynchronously enrich the session record with current weather at the location (can be made up, you don't have to call a weather API unless you want to)
- Every 10 data points, asynchronously calculate the total distance and average velocity, updating the session record with the results
- Build a webpage that displays the route taken for the session on a world map
- Each data point should post a real-time notification to the map webpage that includes the latitude and longitude
With the requirements above, an application would implement:
- A CRUDL API
- An event-driven architecture triggered by API events and data saves
- Complex data structures and indexes/collections
- Real-time notifications
This far exceeds a basic CRUD app that saves and fetches data from a database. A framework that handles the above capabilities would adequately cover many of the requirements of a production-ready application.
Your task is to create an API that ingests data, not creates it. With that in mind, this repository comes geared up with a data generator that will constantly throw data at your API.
The data-generator script will create random GPS data along a route somewhere in the United States. It posts the data points along a route to the API you created until you tell it to stop (ctrl+c or command+c in a terminal). Specifically, it will perform the following actions for each data point it generates:
- Wait randomly between 1-5 seconds
- Move north/south/east/west randomly, slightly prioritizing the last direction it went
- Travel at a random speed between 1 and 80 miles per hour
- Send a POST request to
/sessions/{sessionId}/datawith the random latitude and longitude
Before running the script, you must update it with the baseUrl of your API. To do this, update the BASEURL property on line 3 of the script with the base url of your API. You do not need to include /sessions/{sessionId}/data in url!!!
The script accepts a single command line argument for a sessionId. After you create your session, you can execute the script locally. If we imagine our sessionId was allens-session, then the command would look like:
node .\data-generator.js allens-sessionThis script will run until you stop it manually.
We aren't all experts at the UI. Included in this repository is an example html file of a rendered route. Feel free to use it as a basis for your assessment. The example uses OpenStreetMap and Leaflet for route visualization.
Consider using something like React Leaflet for easy map manipulation and Momento Topics for real-time communication.
Go build it! Use this assessment to test any platform or framework of your choosing. Use it to build a standard application that you can objectively compare across multiple platforms.
Good luck and happy coding!!