Geospatially-verified grant applications, written and audited by deterministic AI.
AgGrantDSS takes a real farm parcel, pulls every authoritative spatial layer that describes it, runs the proposed grant project through the funder's eligibility criteria, and produces two artefacts in one go: a grant application that an applicant can submit, and an audit dashboard that the funder can defend.
The journey, end-to-end
A real working farm becomes a defensible grant application in three deterministic steps. Every number on the application traces back to a named, public dataset.
Submit an application
A 6-step wizard captures the applicant, property (Lot/DP), project scope and components, and supporting evidence. The Lot/DP triggers a polygon resolution and intersection against every authoritative NSW spatial layer.
Open wizard →Grant application
The proposed project is matched to the funder's eligible-works list, modelled for productivity, drought and sustainability uplift, and packaged as the artefact the funder asks for.
View application →Audit dashboard
The same data, viewed by the funder's auditor. Every eligibility criterion is checked and either marked as pass, conditional, or routed to a human team — with the citation that justifies the verdict.
Open audit →Why deterministic
Every number cites the dataset it came from.
The platform does not generate text from a language model and hope it is right. It runs polygon intersections, raster reads and structured rule checks against authoritative public datasets, then writes the report from the results.
The applicant gets a defendable submission. The funder gets a defendable approval — and a paper trail that survives audit.