From Vets to VET: Lessons from WorkMate’s AI model for Veterinarians

Overview

Artificial Intelligence (AI) is transforming industries worldwide — and education is no exception. While previous articles in this series have explored AI’s growing role in learning, this edition steps outside the classroom to ask: what can education learn from AI tools designed for completely different sectors?

To explore this, we interviewed veterinary data scientist Tom Brownlie, founder of veterinary AI tool WorkMate. His journey, especially given the advanced state of his product, offers valuable insights for education professionals looking to develop or adopt AI solutions.

This article is the eighth in a series titled “AI in Education”, aimed at education providers interested in AI. The intention is for this series to act as a beginner’s guide to the use of AI in education, with a particular focus on AI agents. This series is being developed as part of a project to develop an AI agent for learner oral assessment, funded by the Food and Fibre Centre of Vocational Excellence (FFCoVE). We invite you to follow along as we (Scarlatti) document our learnings about this exciting space.

Meet Tom Brownlie: From (almost) farmer to founder

Tom Brownlie describes himself as a “wannabe farmer” who now works as an epidemiologist in the heart of Otago. In 2016, he and co-founder Christopher Laing created Ingenum, a veterinary data science company focused on disease surveillance.

On screen, Tom has a big presence — but he’s instantly comforting. Humble, warm, and always ready with a well-timed joke, he makes you feel at ease straight away. His career spans education and practice — including time as a lecturer at Massey University, a Rural Professional Services Manager at LIC, and a practicing vet.

As the company grew, Tom and his team turned to AI to better detect disease outbreaks and reduce veterinarians’ admin burden. This led to the development of WorkMate — a smart AI tool designed to support frontline vets.

The birth of WorkMate: AI for vets

WorkMate is designed to support vets throughout their entire workflow — from the moment they finish a consultation to generating client summaries and invoices. Here’s how it works:

Input: The vet speaks or types a summary of the appointment. This supports fast-paced work and vets with learning differences like dyslexia.

AI processing: WorkMate connects to clinical literature and a database of anonymised case histories to:

Summarise the animal’s clinical history
Answer clinical questions
Suggest treatment options
Flag gaps or inconsistencies in notes.

Output: The tool can then generate:

A clinical summary
A treatment plan
An invoice
Care instructions for the animal’s owner.

This reduces admin time and enhances clinical decisions. If used widely, Workmate could even track the spread of diseases.

Why should VET educators care?

WorkMate is designed for veterinarians. At first glance, those in VET may wonder, why should an AI tool for veterinarians matter to us? We argue this is because WorkMate is an ideal case to learn from - it shares many technical similarities to an AI agent for oral assessment, plus it is a unique example of a mature AI product built in New Zealand.

Here’s how the two tools compare:

Tom’s tips for building AI tools

At the end of the interview, we asked Tom what advice he would give people thinking about developing an AI agent for use in education:

Don’t start from scratch
“Building an AI tool isn’t something to undertake casually”
Explore existing tools before building your own. Many needs can be met by customising existing models (e.g., adding a ‘skin’) rather than starting fresh.

Let your users lead
“Your product development must be led by your target market”
Talk to teachers, learners, and providers early to ensure features solve top-of-mind problems. Features like WorkMate’s invoicing and note generation came directly from user feedback.

Build for flexibility
“Make it so it can be adapted to the users”
You will want tools that can be localised or customised — whether for different regions, subjects, or assessment types. This needs to influence the build from the very beginning.

Test with the sceptics and influential people
“Test your product on people who are the least likely to use the product”
Invite sceptical or tech-averse users to trial the tool. Also invite those who may influence their organisation’s uptake. Their insights — and adoption — may make or break broader uptake.

Demo, demo, demo
“Cut through the noise to get it seen”
It can be hard to get users to try something new if they feel existing methods are working. Demonstrations are powerful. Seeing the tool in action — at conferences, meetings, or live sessions — builds understanding and trust.

Scarlatti’s take

Just because an AI tool is created with one specific sector in mind does not mean that its underlying technology or the lessons learnt by its founders can't be applied to another area. As AI expands and develops across different sectors, we can continue to gain valuable insights into the opportunities and potential risks that it poses by learning from others.

Questions that we are asking for our own AI agent:

How can we continue to learn from tools outside of education?
What design lessons apply across sectors?

Interested in following our journey into AI?

Sign up to receive our next article directly to your inbox.
Contact the Scarlatti team to share your thoughts or questions.

From Vets to VET: Lessons from WorkMate’s AI model for Veterinarians

Phoebe Gill

Overview

Meet Tom Brownlie: From (almost) farmer to founder

The birth of WorkMate: AI for vets

Why should VET educators care?

Tom’s tips for building AI tools

Scarlatti’s take

You may also like

Cogniti AI: Choosing a platform or custom agent

Māori perspectives on AI: Researchers and Innovators

Government guidance: For educators using AI agents