Flash Decision - an application designed to enhance your securities trading skills through comprehensive training.
- Backend: Python 3.11
- Frontend: Vue.js 3
- SQL Database: PostgreSQL
- NoSQL Database: MongoDb
- Cache: Redis
- Observability: Grafana & Loki
- CI/CD: Docker containers & GitHub Actions
- Built with a focus on high maintainability, scalability, and testability, utilizing architectural patterns like Clean Architecture and Domain-Driven Design (DDD).
- Employs a modern technology stack: FastAPI, SQLAlchemy 2.0 with async ORM, Pandas 2.0 and Pydantic 2.0.
- Implements powerful logging using
structlogandrich. JSON-based logs for production and aesthetically pleasing RICH logs for development. - Offers database flexibility through
alembicdatabase migrations, with two separate schemas for production and development data. - Embraces code conciseness using the
attrsandpydanticlibraries. - Ensures high code cohesion and low coupling by applying the dependency inversion principle and employing various software design patterns:
- Classic patterns: Factory, Singleton, Command, State, Observer, Facade.
- DDD patterns: ValueObject, Aggregate, Repository, UnitOfWork.
- Enhances code quality by utilizing rich types through Generics and other techniques.
- Simplifies high-level configuration with the
BootstrapandConfigmodules, along with the use of environment variables via dot-env. - Maintains a high-quality codebase with linters such as
mypy,ruff,black, andisort. - Minimizes low-quality code commits with
pre-commithooks. - Ensures code reliability with extensive testing, covering 94% of the codebase through a combination of 98 integration and unit tests, evaluated using
pytest-cov.
- Over 100 tickets on the project's Backend Kanban board and project's Frontend Kanban board
- More than 180 uniform Git commits with detailed descriptions that are linked to Kanban tickets.
- The app can operate in 2 modes: production or development.
- A useful Makefile is included to automate and streamline the build process.