Gig Worker Earnings Tracker

A lightweight, AI-assisted Flask web app for delivery analytics — built because spreadsheets were slowing me down.

Overview

When DoorDash became my side-gig, I tracked everything in Google Sheets.
But the formulas, sheets, and manual updates became chaotic… so one evening, out of boredom (and curiosity), I challenged myself:

“Can I replace the entire spreadsheet workflow with a clean local web app?”

Three nights later, I shipped DoorDash Earnings Tracker — a fully functioning Flask + SQLite analytics system, designed to log, analyze, visualize, and export earnings and expenses with zero cloud dependency.

Why I Built It

I needed faster weekly insights without scrolling through formulas.
I wanted beautiful charts that Sheets didn’t give me.
I wanted local-only data: no cloud, no account, no risk.
I wanted to experiment with a production-style Flask project, including authentication, CSRF, modals, systemd deployment, etc.
And mostly…
I was bored and wanted to see how fast I could ship a real app using my AI-assisted coding workflow.

Core Features

1. Dynamic Dashboard

The homepage shows everything instantly with clean KPIs and smart presets.

Date Presets: Week | Month | YTD | All-time
KPIs: Total earnings, Avg/hr, Avg per delivery, Expenses, Net, etc.
Visuals powered by Chart.js:

Earnings over time
Rolling 4-week trend
Category breakdown
Delivery vs. hours ratio

2. Weekly Earnings Management

All weekly logs are editable via modal forms with full CRUD:

Week number, start/end date
Hours worked
Deliveries count
DoorDash Pay + Tips + Other Pay
Out-of-pocket expenses
Notes
Soft-delete & restore support
Filters, sorting, and pagination

3. Expense Tracking System

Separate tables for:

Categories
Individual expenses (with date, amount, notes)
Optional receipt field
Search, category filter, pagination

4. Analytics, Export & Backup Tools

CSV Export: weekly.csv, expenses.csv, summary.csv
Automatic backups to /backups
Restore from UI (admin-only)
Data import via import_data.py with:
- duplicate detection
- dry-run mode
- truncate mode

5. Security & Local Auth

PIN lock with hashing
Session timeout
CSRF tokens
Local-only storage (SQLite)
No internet dependency

Architecture

Backend: Flask + SQLAlchemy
Database: SQLite
Frontend: HTML, Jinja2, Tailwind-style minimal CSS, Chart.js
Auth: Local PIN hash + session timeout
Packaging: Bash scripts + systemd service template
Config: JSON-driven (currency symbol, date range presets, PIN hash)

File Structure

app.py                # routes, models, auth, csrf, exports
templates/
    base.html
    index.html        # dashboard
    weekly.html
    expenses.html
    admin.html
    login.html
static/
    script.js         # charts, filters, modals, theme toggle
    style.css
import_data.py        # CSV importer
config.json
dist/                 # packaged tarball

Screenshots

What I Learned

This wasn’t just a “replace my spreadsheet” project. It became a small production-grade system:

Designing API endpoints for chart JSON
Implementing a real CRUD experience with modals
Managing soft deletes
Packaging with bash automation
Building CSV export logic correctly
Handling CSRF in Flask
Writing custom import scripts
Turning quick ideas into polished tools

The “AI Vibe Coding” Workflow

This project was intentionally built fast using an experimental workflow using Codex and My Curiosity:

I described the app at a high level.
AI generated the first version of each module.
I refined the architecture and debugged local behavior.
Together, we iterated through UI, chart logic, and packaging.
Shipped from idea → working app in a very short cycle.

That speed and creativity became part of the fun.

Conclusion

DoorDash Earnings Tracker is a small but meaningful example of how quickly a well-structured idea can turn into a working product — especially when pairing human creativity with AI-accelerated development.

It replaced spreadsheets that I use completely.
It gave me cleaner insights.
And it proved to me that I can ship real tools, fast, while working on networking, labs, and everything else in my busy routine.