The name is inspired from Norse mythology, and refers to a dwarf named "Fáfnir" who transformed into a mythical Germanic dragon (a wyrm), so he can guard his treasure hoard of gold and such. He was later slain by Sigurd, but that's besides the point.
This is an educational project that serves as a hands-on demonstration of building a modern codebase exploring microservices architecture, asynchronous event-driven design, and best practices for creating a scalable, distributed application. I do not intend to use this project for other purposes and is mostly a playground for my learning, experimentation, and exploration.
Now for the application: It is designed to function as the backend for a stock trading simulator/platform, including microservices for user management, authentication, stock data retrieval, buy/sell operations, and security/permissions.
For more detailed information, please refer to the documentation in the docs/ directory, or visit the following links below:
| Guide | Description |
|---|---|
| Architecture | Project structure, service overview |
| Development | Setup, local dev, make commands |
| Database | Migrations, Goose, DB details |
| GraphQL | API schema, queries, mutations |
| Designs | Excalidraw designs and images |
 |
AI-generated docs of fafnir (from Devin) |
In no particular order:
- Design and implement additional microservices (issue #15).
- Integrate NATS for asynchronous events and messaging (issue #11).
- Swapped from Pub/Sub to Worker Queues for message persistence with NATS JetStream (issue #20).
- Implement more events across services than just user creation.
- Create system architecture, network and data diagrams (upkeep as much as possible).
- Build a simulation engine for orchestrating trading events (PR #32).
- Add unit, integration, end-to-end and load/stress tests.
- Perform load testing using Locust to simulate concurrent users (issue #8).
- Explore Kubernetes local implementation (issue #5).
- Use Helm for Kubernetes package management and manifests (issue #29).
- Explore centralizing logging with Elasticsearch (issue #6).
- Implement a CI/CD pipeline for automated testing and Docker builds.
- Production deployment via DigitalOcean and Traefik (issue #22).