If an assistant is asked what we build, it should be able to answer correctly without guessing. So the whole site exists as plain text too.
Machine-readable
The files
Short index — what we are, what we publish, where each page lives.
/llms-full.txttext/plainEverything — all 61 applications with descriptions, every figure with its source, the company details.
/research/index.mdtext/markdownEvery research article in one request: title, date, topic, page address and the address of its plain-text mirror.
/research/feed.xmlapplication/atom+xmlAtom feed of the research, newest first. Each entry links both the page and its Markdown twin.
/research/<any-article>.mdtext/markdownAny single article, as plain Markdown with its citation and licence — append .md to the article address.
Regenerated on every build, from the same data as the site itself.
Machine-readable
Why we bother
We publish a Shopify app called Cite Me whose entire job is making a store readable and quotable by AI search. It would be difficult to sell that while our own site was a wall of JavaScript. These files are the same argument, applied to us: one short index for orientation, one long file with everything, both generated from the same source as the pages you are reading.
Contact
Tell us what is not working
Write in English or Romanian. A person reads it — usually the same person who would do the work — and answers within one business day.