Sell-Through based forecasting for reatil buying

For my final-year project I have to demonstrate a complete sell-through based buying-forecasting system that runs in the browser. I already drafted page layouts and will share the reference UI the moment we start, so you can focus on the build rather than design decisions.

Core scope
• Authentication: single-email login (no social OAuth).
• Dashboard with a left-hand sidebar for easy navigation.
• Data entry screens where sales and stock figures are typed in manually; later we can leave hooks for CSV or API if time permits, but manual entry is the required path for this submission.
• Forecast logic that projects buying requirements from sell-through; I will explain the formula and provide sample datasets.
• Filters to slice results by Season, Range and Channel (EBO / MBO / LFS).
• Dual warehouse tracking for Jigni and GGN.
• Result view presented first as a clean HTML table; optional chart components are welcome if time allows.

Tech stack
React for the UI, Firebase or Node with MySQL for the backend—feel free to pick the combination you can deliver fastest, as long as the code is clean and easy to demo.

Deliverables
1. Source code in a public or private repo.
2. Deployed demo link (Firebase Hosting, Vercel, or similar) ready for my viva.
3. Quick setup guide and README.

Acceptance criteria
– All listed features reachable from the sidebar without page reloads.
– Forecast output matches my sample calculations within ±1%.
– Code passes a brief walkthrough in front of faculty (clear comments, no hard-coded credentials).
– Completed and handed over inside 7 days so I have buffer time for testing.

If you can move quickly while keeping the interface simple and the logic accurate, this project will be straightforward and a nice portfolio piece for both of us.