Expense Tracker (Work in Progress) 🛠️
This is not a finished project.
This is something I am actively building to improve how I think about systems, structure, and real-world application design.
Repo: Link will be provided when i complete my learning
What this is
This project is my attempt to move beyond small scripts and start building something that evolves over time.
Instead of jumping between different ideas, I chose one problem — an expense tracker — and started rebuilding it multiple times with better structure and better decisions.
Right now, the repo contains multiple stages: - A basic single-file script - A modular CLI version - An argparse-based CLI - An early FastAPI backend
Each stage represents how my thinking changed, not just how the code changed.
What I am trying to learn
Right now, the focus is not features — it’s structure.
- How to break a program into responsibilities
- How to move from scripts → tools → systems
- How to design something that can grow without becoming messy
- How real applications are structured (not just how they run)
The shift I’m working on is:
from writing code that works → to designing systems that make sense
Current state
- CLI versions are functional and usable
- Argparse version feels closer to a real tool (but no one will use it, who uses a cli expense tracker)
- FastAPI version is still early and incomplete
The API stage is where things are starting to feel different — exposing logic through endpoints instead of directly running functions forced me to rethink how everything is structured.
What I’ve realized so far
- Writing working code is easy compared to organizing it properly
- Most complexity comes from managing state and structure, not features
- Rebuilding the same project teaches more than starting new ones
- My earlier versions worked, but didn’t scale or feel clean
What I’m focusing on next
- Improving the FastAPI layer
- Separating business logic from API routes
- Introducing proper data handling (moving beyond simple JSON)
- Preparing for a database-backed version
What this project is NOT (yet)
- Not production-ready
- No authentication
- No database design
- No deployment
That will come later — right now I’m building the foundation.
Why I’m doing this
I realized that jumping between projects made me feel like I was progressing, but my thinking wasn’t improving enough.
So I changed the approach:
One idea. Multiple implementations. Increasing depth.
Final note
This project is intentionally unfinished.
It exists to show how my thinking is changing, not just what I’ve built.