Bank alerts to a day-book, with a local Gemma

Description and use

A browser‑based day‑book app for a Nigerian POS operator uses Gemma 3‑1B to extract customer names from bank transfer alerts, running offline on the user’s GPU and eliminating manual notebook entry.

Problem

Mama POS needed a way to record bank transfer alerts without manual note‑taking, to avoid errors and reduce time.

Outcome

The app correctly extracted 5/5 kinds and 4/5 names from real alerts, and 30/30 kinds and 23/30 names from synthetic alerts. It runs offline after a one‑time 859 MB download and requires a GPU for the full model.

Workflow

• Open the day‑book app in the browser
• Paste the email alert into the app
• Code extracts time, amount, and kind of money
• Gemma 3‑1B reads the customer name from the alert text
• App displays a row with all fields and a status (Landed/Not seen yet/2 payments)
• Repeat for each alert
• At end of day, app shows totals, estimated charges, rows needing review, and a WhatsApp‑ready text
• Copy the text to WhatsApp and send to business owner

Evidence

She writes each payment in a notebook, and it "takes long".

Source

All data stays in the browser. Nothing goes to a server.

Source

Gemma 3 1B runs in the browser on the user's own GPU and does the one job in the app that code cannot do: reading the customer's name from free text written differently by every bank.

Source