Gemma Garage: a local Gemma reads a mechanic's scribbled notes and orders the parts

Description and use

Gemma Garage automates a car workshop’s workflow by locally interpreting free‑text notes with Gemma3, ordering parts the day before, sending Telegram reminders for repeat jobs, and generating invoices, all while keeping customer data on the local machine.

Problem

A car workshop owner keeps job notes as scribbles and manually orders parts, tracks repeat jobs, and generates invoices, leading to inefficiencies and data privacy concerns.

Outcome

Gemma Garage streamlines workshop operations by automating note parsing, part ordering, repeat job reminders, and invoicing while keeping all customer data local, reducing manual effort and privacy risks.

Workflow

• 1. Enter job note in Spanish (e.g., "cambio aceite motor opel aceite 5w30").
• 2. Local Gemma model (gemma3 via Ollama) parses note into job, car, parts, and repeat frequency.
• 3. System displays live interpretation for review.
• 4. Business day before appointment, system orders parts from distributor; user can review/edit order; failed orders are flagged for retry.
• 5. System groups parts into a shopping list for the day.
• 6. For repeat jobs, system sends a Telegram reminder to the customer one week before the yearly deadline.
• 7. After job completion, user enters part prices from delivery note and labor tariffs; system calculates total and generates an invoice with VAT.
• 8. All data remains local; only distributor order and optional Telegram bot communicate externally.

Evidence

A small appointment book for a relative who runs a car workshop. He keeps his jobs as scribbles like cambio aceite motor opel aceite 5w30

Source

Gemma Garage keeps the scribbles and does the paperwork: ... A local Gemma model reads it and shows, live, what it understood: the job, the car, the parts and how often it repeats.

Source

The business day before the appointment, the parts are ordered from the distributor (Monday for Tuesday, Friday for Monday). He can review and edit the order first, and failed orders are flagged to retry.

Source

Repeating jobs trigger a Telegram reminder to the customer a week before the year is up.

Source

When the job is done he types the part prices from the delivery note (they are never stored, because they depend on the distributor), labour comes from his own tariffs, and an invoice with VAT is generated.

Source

Nothing leaves the machine. Customer names, plates and phone numbers never go to a cloud API. The only outside calls are the distributor order and the Telegram bot he opts into.

Source