This VFS implentation is an in-memory VFS shipped with its REST API and a web client.
DISCLAIMER: This project is for demo purposes, it is not intented to be used in production.
I've developed many things throughout my career, but I've realized that I can't showcase much of it. Most of the projects I've worked on are private. One day, I asked myself: How can people evaluate my work if they can't see what I'm capable of? That's why I built this VFS.
For those who are curious and reading this, I will guide you through the application, demonstrating my skills—or at least a subset of them.
I hope that you will enjoy to read my code.
The project is composed of three parts: the vfs module, the server, and the client. The backend is developed in Go, while the frontend is developed in Next.js.
The vfs module is the database engine. It’s a tree of nodes representing the directories and files of the filesystem. It provides convenient functions to manipulate files. Learn more here
The server is a Go REST API developed with gin. I built it following the principles of screaming architecture. The server allows manipulation of the VFS via HTTP. Learn more here
The client is developed with Next.js. There’s nothing particularly unique about it—it simply allows browsing and manipulating your files. here
The product allows you do to the following things:
- Upload files
- Create directories
- Delete files or directories
- Nest files and directories
- Renaming files and directories
- Download files
Dockerinstalled on your machine
To run the project, clone it and run docker compose up command in a terminal at the root of the project.
To access the VFS, open a web browser and go to http://localhost:3000.
I chose to organize the project in a monorepo for simplicity and ease of demonstration. It makes navigation more straightforward. However, in a production environment, I would split the VFS, server, and client into separate repositories. This approach provides clearer versioning and management of each component individually.
For the backend, I have over 4 years of experience developing backend applications in Go, primarily focusing on microservices. As a dedicated Gopher, I believe Go is one of the best languages for building backend applications.
For the frontend, although I love React and also appreciate Vue.js (Nuxt), I chose to use Next.js. In my opinion, Next.js allows for faster development compared to Nuxt, even though Nuxt offers more structure to the code.
The eternal war between backend and frontend engineers. Based on my previous experiences, I would say backend. It is my specialty and where I am most proficient. However, I enjoy taking breaks to work on frontend as well. I like frontend too. I would position myself as a backend-oriented fullstack engineer (if that’s a valid term).
As you’ve seen, I know how to use Docker. To deploy real applications to the world, I am familiar with using a Kubernetes cluster and deploying apps on it using GitOps principles. Additionally, when a Kubernetes cluster is not needed, I prefer using PaaS platforms such as Vercel or Heroku.
For Kubernetes, here is what I do:
- Use Flux for continuous delivery and GitOps
- Secret encryption with SOPS
- Monitoring, alerting, and logging stack: Prometheus, Alloy, Loki, and a Grafana instance
- Security scans: Kubescape
- Ingress: Traefik, LetsEncrypt, CertManager
This stack has been used in production and works well.