Skip to content
Tejobit

AI

Gemini integration

Google's Gemini models reading photos, receipts and scanned documents, working on data from Gmail, Drive and Sheets.

Request this integration See all integrations

Overview

Gemini

Gemini models are multimodal by design: they read text, images, PDFs, audio and video in the same request. For businesses that get a lot of information as photos (expense receipts, delivery notes, site photos, meter readings), that removes a lot of manual typing.

Tejobit connects the Gemini API to the company's Google Workspace. The connection fetches files from Drive or attachments from Gmail through Google's APIs, sends them to Gemini with clear instructions and writes the result where it is useful: a Google Sheets row with supplier, tax ID, date and amount, a summary in a doc, a label on the email.

One point on data matters: under the Gemini API terms, free usage may be used by Google to improve its products, while paid usage is not and falls under the data processing addendum. The connection therefore always runs on a billing-enabled project. For EU data location requirements, Vertex AI is the alternative.

Use cases

What it is for

  • A building contractor whose crews email receipt photos and get expenses logged in a sheet.
  • A distributor reading scanned delivery notes and checking quantities against the order.
  • A property manager extracting meter readings from photos.
  • A weekly summary of emails under a Gmail label, with pending requests and deadlines.
  • Product descriptions for an online shop drafted from photos, for review.
IN

What the connection listens to

  • New Gmail attachments under a set label or address.
  • New files in a Google Drive folder, including photos taken on a phone.
  • New Google Sheets rows or scheduled summary jobs.
OUT

What the connection does

  • Read images and scanned PDFs and extract supplier, tax ID, date, total and VAT.
  • Write extracted data to Google Sheets or another system, linked to the original file.
  • Summarise email threads and Drive documents.
  • Describe or classify images, for example site photos by stage or issue type.
  • Flag low-confidence results for a manual check.

How it works

From the request to a working connection

  1. 01

    Request

    The form names the integration, the tools involved and what should happen in each case.

  2. 02

    Review and proposal

    Tejobit confirms what the API allows and sends a proposal with the flow, the price and the timeline.

  3. 03

    Setup and testing

    The connection is set up with the business's own accounts and tested with real cases before it goes live.

  4. 04

    Maintenance and monitoring

    Tejobit hosts the connection, follows up on errors and adapts it when the API changes.

The connection runs on Tejobit's infrastructure. The monthly fee covers hosting, monitoring and API changes.

What to prepare

  • A Google Cloud project with billing enabled for the Gemini API (usage is paid to Google).
  • Authorised access to the Google Workspace accounts, labels or folders involved.
  • Real samples of receipts, delivery notes or documents, including the hard-to-read ones.
  • Rules on which personal data may be processed, in line with GDPR.

For security, do not send passwords, API keys or tokens through the form. Credentials are requested through a secure channel when the setup starts.

Limits

  • Blurry or dark photos and faded thermal paper give incomplete readings; the connection flags them for review rather than guessing.
  • Receipts and invoices read by AI are for logging and checking; they do not replace the original document for tax purposes.
  • Free Gemini API usage may be used by Google, so only the paid tier is used.
  • Per-minute request limits and per-token cost depend on the model and account tier.

Frequently asked questions

Before you start

Is this the same as Gemini inside Google Workspace?

No. Gemini in Gmail or Docs is a per-user assistant. This connection uses the Gemini API to process files and email automatically, by set rules, and write the results into other systems.

Can it read thermal paper receipts?

Usually yes, as long as the photo is sharp and the receipt has not faded too much. Anything it cannot read with confidence lands in a list for manual checking.

Does Google train on the data sent?

With billing enabled, the current Gemini API terms say prompts and responses are not used to improve Google's products. Free usage may be. Check the wording in force; the connection always runs on the paid tier.

Why Gemini rather than ChatGPT or Claude?

When the raw material is images, scans or files already in Google Workspace. For high-volume text extraction OpenAI is an alternative; for very long documents and careful writing, Claude.

Related integrations

See all integrations

Set up the Gemini integration?

Describe what should pass between the tools. The reply confirms feasibility, price and timeline.

Talk about your project geral@tejobit.com