Objective

The objective of this assignment is to assess your Backend development skills, problem-solving abilities, and code quality. You will be creating a secure and scalable backend service built on Node.js and Express.js. Similar to a minimal version of AnswersAi, this service will take in user requests and handle questions. This is designed to be used with a frontend application - and while you do not need to build any frontend, please keep this in mind while designing this service.

This assignment has been designed to be doable within 4-5 hours. You will have 5 days from when you receive the email to submit this assignment. If you need more time, please email [email protected] to request an extension.

You do not need to complete 100% of task requirements. Please complete as much as you can within the 5-day timeline - we value speed over perfection.

Requirements

1. Design a RESTful API using Node.js and Express.js with the following endpoints:

   • POST /api/questions
     • Accept user question, and return AI-generated answer.
   • GET /api/questions/:questionId
     • Retrieve specific question and answer by question ID.
   • POST /api/users
     • Create a new user account.
   • GET /api/users/:userId
     • Retrieve a user profile with a given userId
   • GET /api/users/:userId/questions
     • Retrieve all questions asked by user with a given userId

2. Implement authentication and authorization using JWT tokens or similar for API security.

   • POST /api/auth/login: User login endpoint.
   • POST /api/auth/logout: User logout endpoint.
   • POST /api/auth/refresh: Refresh access token endpoint.

3. Design and implement a database schema for questions and users.

   1. Ensure that this is scalable, and be prepared to explain your design choice (relation vs non-relational, ACID guarantees, etc).
   2. Keep in mind that this service will scale to hundreds of thousands of users who might use it concurrently.
   3. Optional: Use an ORM (such as Sequelize or Knex.js).

4. Integrate with an AI service (e.g. OpenAI API) to generate answers. Use secure libraries/SDKs for API calls. We recommend langchain.

   1. Keep in mind that this service will scale to hundreds of thousands of users who might use it concurrently.
   2. Please document a solution (no need to code it) that ensures that users get timely responses even during periods of high load.

5. Containerize the application using Docker. Provide a Dockerfile for easy deployment.

6. Design a scalable infrastructure on AWS or GCP (EC2, RDS, ELB, ECS, Auto Scaling, etc.). Provide an architecture diagram or description.

   1. You do not need to deploy to the cloud. Simply provide an architecture diagram (using software like Lucidchart or on pen and paper), or a written description.

7. Write unit tests for critical components (API endpoints, database interactions) using Jest, Mocha, or similar libraries.

8. Provide a README with setup and running instructions, including dockerfile setup and running instructions, database setup, environment variables, and dependencies.

Evaluation Criteria

Your assignment will be evaluated based on the following criteria:

• API Design and Implementation (20%):
  • Proper design and structure of API endpoints
  • Adherence to RESTful principles
  • Appropriate use of HTTP methods and status codes
  • Clear and consistent API documentation
• Database Design and Interaction (20%):
  • Proper database schema design
  • Appropriate use of data types and constraints
  • Efficient and optimized database queries
  • Proper handling of database interactions (CRUD operations)
• Authentication and Authorization (15%):
  • Secure implementation of user authentication using JWT tokens
  • Proper handling of token creation, verification, and expiration
  • Secure storage and handling of sensitive information (e.g., passwords)
• Code Quality and Best Practices (15%):
  • Adherence to coding best practices and conventions
  • Proper code organization and modularization
  • Appropriate use of error handling and logging
  • Efficient and optimized code implementation