Skip to main content
Provenance & defensibility

Every number says what produced it

Content-addressed results, manifests that record the kernel, inputs, parameters, actor and toolchain, and structured refusals instead of plausible-looking answers — so a technical review can verify rather than trust.

Request DemoView all solutions
The artifact inspector — what ran, on what inputs, with which parameters, by whom, and under which toolchain pins, read from the artifact's own manifest.

Target audience

Competent Persons, technical reviewers, auditors, due-diligence teams, QA/QC leads, and regulators.

Platform capabilities

Content-addressed artifacts — the same job returns the same result identity
Manifests recording kernel digest, input identities, parameter hash, randomness root, and actor
Lineage graphs walked from manifests alone
Units, support, and epistemic class declared per attribute
Source database, tables, analyte, and drilling recorded at capture
Structured refusals with code, category, and stage — never prose to parse
Stored results re-verified on every read

Business benefits

Shorten technical review by making claims checkable
Survive due diligence with evidence rather than recollection
Catch invalid work at submission instead of in the report
Keep interpretation, computation, and classification clearly separated

Screenshot gallery

Reproducible Geoscience & Audit in the product

The screens that support this workflow, shown in the working application.

Provenance that reaches back to the database: the tables read, the canonical project, the analyte, and the drilling behind the estimate — labelled as capture evidence, not a claim the kernel makes.
Refusals cross the boundary as structure — code, category, stage, kernel — so an invalid job is explained and actionable rather than answered with an invented number.
The artifact store — every persisted result retrievable by its content address, re-verified on every read.
The job builder for ordinary kriging — assumption parameters are required, have no pre-filled default, and say why: a silent default is a decision nobody made.

Compare this workflow with the full platform

Explore adjacent capabilities in AI agents, data management, remote sensing, 3D visualization, and enterprise GIS.

View Features

Request demo

Request a Reproducible Geoscience & Audit demo

Bring your workflow, target users, and data readiness questions. The demo can be tailored to the product areas shown on this page.

Request DemoExplore Platform