This is my submission to a competition to work at Diligent as a Senior Software Engineer.

Try it out live! Deployed here: https://diligent-assessment.schwarzkopfb.codes
Task description: https://github.com/dil-ajanek/tech_assessment/blob/main/README.md

This repository consists of 3 main folders and some configuration and manifest files.

• prisma dir contains the database schema and initial "migration" sql file for a PostgreSQL datastore - used by the backend.
• public is the classic "wwwroot", containing static assets served directly and a special index.html which is used as a template for the final client html.
• in src there are two subdirectories: client and server containing the source files for the frontend and the backend respectively.

The client is based on React with TypeScript, Bootstrap and Sass for styling - bundled by webpack.

The server is also written in TypeScript on top of Koa because that's a quite powerful, light-weight web framework with amazing performance and flexibility.

The project is backed by PostgreSQL. Prisma ORM is chosen because of it's compatibility with AWS Aurora and the easy, reliable management of migrations as well as its superior development experience with Node.js.

First install the dependencies listed in package.json:

$ npm install

Then build both the server and the client code:

$ npm run build

this will emit the transpiled and bundled client and server code into the dist folder.

To be able to connect to The Movie Database, you'll need an access token for their API and also a PostgreSQL database should be available somewhere. You have to set DATABASE_URL and TMDB_API_READ_ACCESS_TOKEN in the shell you're going to run the application.

Once the build step is done, and the configuration is ready, you can start it up by executing:

$ npm start

A minimal but useful development logger is also included.

You can override the default port 3000 with the PORT environment variable. But if all the above steps succeeded, now your server is ready to accept connections at http://0.0.0.0:3000/.

If you're planning to actively develop the project there is also

$ npm run dev

which starts up incremental compilers with watchers, so you can try out your work locally right after saving file changes in the project.

Due to the standard structure of the project, it's directly deployable to Cloud providers like Heroku and easily adjustable to Serverless hosting providers to be able to cost efficiently run it on the edge.

Scalability and performance were the main architectural approaches, as you can experience it yourself on the live deployment - that's powered by a single Heroku Eco Dyno only but is quite responsive and fast.

I tried to a bit over-comment the source to help the reviewer go through it faster.

Hope you'll like it! 🤞 ✌️

MIT