Who Builds This, And What Each Brings
A knowledge platform is worth exactly as much as the people willing to correct it.
Three groups make V.E.T.S. work: the framework partner whose software it is built on, the domain experts whose corrections are the actual product, and the AI layer that writes the first draft so nobody has to face a blank page. This page describes what each contributes, what a contribution technically is, and what the ecosystem does not have yet.
The framework underneath
V.E.T.S. is not built from scratch. It runs on Appframe, an enterprise application framework that supplies the data-binding, the tree structures and the multi-tenant scaffolding that the veterinary logic sits on top of.
There is a checkable way to see how deep that dependency runs. The platform’s own AI knowledge base contains 442 indexed Appframe documents, sitting in the same retrieval corpus as everything else. The framework is not a library that was linked once and forgotten; it is documented, indexed and queried alongside the veterinary material, because the people building on it need to ask questions about it constantly.
What the framework layer actually provides →
Six disciplines looking at one animal
Animal care is not one profession. Each specialist sees the same animal from a different angle, and each keeps their own records in their own system.
Veterinarians
Diagnostics, treatment protocols, surgery, pharmacology.
Trainers
Behaviour, training method, performance over time.
Farriers
Hoof care, corrective shoeing, lameness evaluation.
Breeders
Genetics, pedigree, reproductive management, foal development.
Livestock managers
Herd health, nutrition programs, facilities, scale.
Nutritionists
Feed formulation, requirements by species and life stage, metabolic conditions.
The cross-domain advantage. When a veterinarian’s treatment notes sit alongside a trainer’s behavioral observations and a farrier’s shoeing records, patterns appear that no single expert could see. A horse showing a decline in performance may have a lameness issue visible in the farrier’s notes months before it becomes clinically obvious. Nobody was hiding that information. It was simply in a different building.
Read what these disciplines do with it →
What a contribution actually is
Contributing to V.E.T.S. is not a separate activity with its own workflow. There is no submission form and no review queue to join. A contribution is an edit to a record you were going to make anyway.
The commentary stream is the easiest part to check. The platform holds 4,041 comments, the most recent written on 2026-07-26. That is not a launch-day import; it is what a curation habit looks like after it has been running for a while.
What happens to an edit afterwards runs on two separate schedules, and they are worth keeping distinct:
Every hour
ajob_AI_GCPVectors carries new and changed content into the vector index. Your edit becomes retrievable on its next pass, not the next morning.
Once at 02:00
ajob_AI_Intelligence_Pipeline_Daily does the heavier work in 29 steps across eight phases: summaries, keywords, relationships between documents, and confidence recalibration.
It is worth being precise about what you are contributing to. The retrieval corpus holds 2,333 documents, and all of them are documentation and schema metadata: help text, knowledge-base articles, and generated descriptions of database objects. No animal records, no client details and no patient history are embedded in it. Your corrections to shared knowledge are shared. Your animals are not.
How a correction reaches the retrieval layer →
What the ecosystem does not have yet
The honest description of the community today is that it is small and the tooling for it is thin. There is no third-party developer API. There is no knowledge marketplace, no contributor directory, and no public forum run by the project. Those are roadmap items rather than shipped features, and a page about community that implied otherwise would be the wrong kind of page.
What exists instead is a working platform, an indexed corpus, and a correction path that takes effect within the hour. That is a smaller claim, and it is true.
When the API and the marketplace arrive →
The commercial reasoning behind opening the platform this way — a $12 million five-year plan and four income streams, published rather than pitched — is set out separately.
The business case for an open ecosystem →
Where to go next
If your expertise is in one of the six disciplines above, the useful next step is the reading order for your first week rather than another page about the idea.
The reading order for your first week →
Or go straight to creating an account.