Mock AI Interviewer

• This project is a mock AI interviewer that asks you questions and records your answers 🧠

• The project is split into 3 parts, each with their own respective READMEs:

  • The backend which is responsible for the AI and the database - README
  • The frontend which is responsible for the UI and the user interaction - README
  • The Database which stores the interview types and the conversation history - README

Running locally

• The following section will show you how you can run the project locally

Prerequisites

• Install Docker and Docker Compose (you can just install Docker Desktop for Windows or Mac)
• If you want to explore the mongo database, install MongoDB Compass

Setting up Environment

• Create .env file in ./frontend with the content as shown in the .env.example file:
• Also create a .env file in the ./backend with the content as shown in the .env.example file:
• Fill in missing values with appropriate values

Running the project Locally

• To run the project a combination of different scripts are used

• Scripts that reference docker compose files are run from the root directory

• Other helper scripts can be found in the ./scripts directory

• Currently there are 3 progressive ways to run the project

• To run with hot reloading:

  • Doing it this way will allow you to make changes to the code and see the changes reflected in the browser
  • Run ./run-local.sh and stand up the mongo service
  • Go into the ./backend directory and run uvicorn with the following command: uvicorn main:app --reload
  • Go into the ./frontend directory and run the following command: npm start

• Run project via building containers Locally:

  • Run ./run-local.sh and stand up all services
  • This utilises the local docker-compose file (docker-compose.local.yml)

• Run project via the latest hosted containers on ECR (see section below for more details):

  • RUn ./run-test.sh and stand up all services
  • This utilises the test docker-compose file (docker-compose.yml)

Building Container Images & Pushing to Elastic Container Registry (ECR)

• To do this run the Build and Push Containers github action.
• To find out how to trigger this action look at the on: section of the build-and-push-containers.yml file in the .github/workflows directory
• It should tell you the triggers for the build to be run. ie if it has the push trigger, then pushing the branch will trigger the build
• If it has the workflow_event trigger, then you can trigger the build manually by going to the actions tab on github and running from there. See here

Running on AWS via Elastic Beanstalk (EB)

Deploying to Elastic Beanstalk (EB)

• To do this run the Deploy to Elastic Beanstalk github action
• To find out how to trigger this action look at the on: section of the build-and-push-containers.yml file in the .github/workflows directory
• It should tell you the triggers for the build to be run. ie if it has the push trigger, then pushing the branch will trigger the build
• If it has the workflow_event trigger, then you can trigger the build manually by going to the actions tab on github and running from there. See here