Exploration projects inherit decades of reports, logs and assay certificates. Geology AI can read all of them, but only if every collar, interval and grade it extracts comes with the page it came from.
All Mining & Resources →
A project acquired today typically comes with work by several previous explorers: annual technical reports to the regulator, open-file data from the geological survey, drill logs typed or handwritten, assay certificates from laboratories that no longer exist, and maps in local grids. The information that would change a drilling programme is often in there. Reading it all by hand takes a geologist months.
AI mineral exploration support, as we build it, is mostly careful extraction. Collar coordinates with their datum, hole depths and orientations, lithology and alteration intervals, assay results with their laboratory and method, and the QA/QC that came with them. Each value goes into a structured table with a citation, so the geologist can check any figure in a click and trust none of them blindly.
technical reports, logs, certificates, maps
OCR and vision for scans and tables
collars, surveys, intervals, assays
datum, units, duplicates, QA/QC coverage
to the geological database, cited
conflicts, low-confidence values, gaps
Two ways. Prospectivity models combine geological, geochemical, geophysical and remote sensing data to rank targets. And language and vision models read the historical reports, logs and certificates that projects inherit, turning them into structured, checkable data. The second is often the faster win, because the data already exists and the work is extraction rather than prediction.
Often, with care. Vision models read many handwritten and typed logs well, but legibility varies, so values extracted from handwriting are marked with lower confidence and routed to a geologist for confirmation. Every value keeps its citation to the page image so it can be checked directly.
It can be part of the data a Competent or Qualified Person relies on, provided they are satisfied with its verification. That is why each value carries its source and why QA/QC coverage is reported. Whether the data is suitable for public reporting is the Competent or Qualified Person's judgement, not the system's.
What we built, what stage it is at, and a link to the live thing wherever there is one.
yourOS is an operating system with an AI at its heart. You talk to it like a person, out loud or typed. It makes your calls, plans your days and works on your projects with a crew of agents, on your own private server. Our own product, built from the first commit; the waitlist is open.
Website, backend and app assets for Sonora, a free AI sound-therapy app on iOS and Android published by Cognitive AI.
Brand, website, prototype, apps on five platforms, the AI backend and the algorithm-intelligence pipeline for Podswap, a social growth engine that runs without a human in the loop.
In closed beta: Roxy, a private AI assistant with permanent memory running on infrastructure the user owns. Brand, site, product, backend and conversational design.
Live at coherence.com: brand identity generated with AI agents we directed, the motion system and website for the group behind Sonora, AboutAI, WakeUp and MudShadows. Agent work remains in build.
Live at domainbooth.com for DomainBooth, a premium domain brokerage: a new animated mark, every page redesigned and rebuilt with designed mobile layouts, and a custom CMS for its sales record, featured names, press and enquiries.
Send us one historical report. We will return its drillhole data, with every value cited.
30-minute technical call · no deck