Architecture

Data Models

1. User
   • This table will store user information
2. Portfolio
   • This table will store portfolio information
   • Currently, it doesn't add much value, but allows for extensibility, such as allowing a user to create multiple portfolios
3. CurrencyAllocation
   • This table will store the cryptocurrencies associated with a particular portfolio
4. Currency
   • This table stores all available currencies in the system
   • This should be mostly static data (name, symbol, etc.)
5. CurrencySnapshot
   • This table stores market data for all currencies.
   • It will contain one row for each currency
   • This table will be updated daily with the latest market data.

Project Structure

• project/
  • Django App
  • REST API
  • Scheduled Celery task that pulls ctypto data from CoinGecko and syncs with database
• ui/
  • React App
• docker-compose.yml
  • Used for spinning up local resources for development
  • Redis (Celery Broker)
  • Celery Beat
  • Celery Worker
  • Postgres
• Dockerfile.celery
  • Dockerfile that defines the container for celery beat and worker processes to run in

Crypto Data

Crypto market data gathered via coingecko's public API (https://api.coingecko.com/api)

Improvements

1. Improve usage of serializers
2. Standardized exception handling / response messages
3. Client is not organized at all, focus was on django backend
   • State management is pretty sloppy, all just thrown into a single context with setters drilled down to components
4. AuthN/AuthZ

How to run

1. clone the repo
2. cd into repo
3. run pipenv install
4. run pipenv shell
5. run docker compose up -d
6. cd into project/
7. run python manage.py migrate
8. cd into ui/
9. run npm install
10. run npm run dev
11. site should be running at http://localhost:1234