Two people.
~60 hours saved every month.
Biotika distributes products from ten suppliers to laboratories in Brazil. Quotes, insurance orders, delivery notes: three repetitive tasks that took ~67 hours a month. I automated all three, without changing their tools.
Time spent, task by task
Time per task. Each one happens about 30 times a month.
Three automations, one database
The request arrives in Gmail
A free-form email from the customer, no fixed format.
AI reads the email
Gemini identifies the customer and the requested products.
The product database answers
Storage and shipping temperatures, weight, EUR and USD prices.
The quote is created in Axonaut
Through the ERP's API. The team reviews, then sends.
Olá, por favor, uma cotação para o cliente Laboratório Exemplo :
| REF-1042 Centrifuge, 24 places | ×1 |
| REF-2210 Rotor 4 × 145 ml | ×2 |
| REF-3307 Reagent 500 µL | ×5 |
| Ref. | Product | Price |
|---|---|---|
| REF-1042 | Centrifuge, 24 placesStorage: room temp. · Shipping: room temp. · 20 kg | $3,550 |
| REF-2210 | Rotor 4 × 145 mlStorage: room temp. · Shipping: room temp. · 1.2 kg | $3,630 |
| REF-3307 | Reagent 500 µLStorage: −20 °C · Shipping: dry ice · 0.1 kg | $1,240 |
Illustrative example, fictitious data.
Every shipment must be insured. The team already worked in Google Sheets, so the dashboard was built there: nothing new to learn. Try it: change the drop-downs, then click "Generate".
| Invoice | Customer | From | To | Mode | Status |
|---|---|---|---|---|---|
| F-0142 | Lab Alpha | — | |||
| F-0141 | Institute Beta | Annecy | Paris CDG | Road | ✓ Order #87 |
| F-0140 | Lab Gamma | New York | Campinas | Air | ✓ Order #86 |
Illustrative example, fictitious data.
Start from the invoice
Products and quantities read from Axonaut.
Product data is fetched
From the same database as quotes: shipping conditions, weight…
Delivery note(s) are generated
One or several, depending on the shipment.
They are attached to the order
Automatically, in Axonaut.
The fastest one to build: it reuses the existing product database. Each building block makes the next one cheaper.
Data first, AI second
Product sheets from ten suppliers were scattered, each in its own format. They were collected automatically, normalised and centralised in a database on Google Cloud.
products in one database
“Adaptable and proactive, Emmanuel took full ownership of his assignment and was a real source of ideas!”
Which repetitive task costs you the most time?
In 30 minutes, we look at where your time goes and what is truly worth automating.
Emmanuel Rossi · Freelance AI Automation Engineer
Engineer · ex-Dassault Systèmes (Boeing programme)
rossiemmanuel1@gmail.com