Markdown’s new challenger: The open source DGML format aims to turn documents into data that AI (and humans) can trust.

Jean Paoli has spent his career creating machine-readable documents – first as a co-inventor of XML, then helping to create the file formats behind Microsoft Office. Now his Kirkland, Wash.-based startup, Docugami, is opening up the technology at the heart of its business, betting it could be a common way to turn documents into data that humans and AI agents can trust.

The company is releasing its technology, called DGML (short for Document Graph Markup Language), under the Apache 2.0, widely used open source license, so that other developers and companies can adopt it.

The idea is to turn it into a shared standard that no one company owns, just as XML became the common foundation for the entire technology industry.

Movement reflects the change in which value is created in AI. Docugami has so far developed its own commercial software that turns unstructured documents into actionable data. It is now betting that there is more value in ensuring that the data is reliable instead.

How does this work: Docugami is partnering with Inveniam, a Detroit company whose software helps big investors keep track of the mountains of paperwork behind real estate and other hard-to-value assets. Inveniam will record the type of digital fingerprint of each piece of DGML data on the blockchain created by Mantra, the crypto company acquired by Inveniam.

That means, for example, that a single fact buried in a 200-page lease — such as the rental amount, a renewal option, or a default clause — can be verified on its own, without revealing the entire document. An investor, auditor, or AI agent can trace it to the page it comes from.

To work with documents, AI systems usually convert them into a simple format first. DGML enters a growing field of competitors in that regard, competing with the popular Markdown format and newer formats such as IBM’s Docling.

Business model: This is a big step for a company the size of Docugami, taking a 30-person startup in a new direction. Paoli gives the industry the expertise his team has spent years shaping, and pinning the company’s future to a bigger vision.

The plan is to make money not on the format itself but on the value of reliable data. Once a company converts a lease or loan into DGML and embeds key numbers into the blockchain, investors, lenders and auditors can pay to draw on that verified data.

Docugami will share profits from its partnership with Inveniam. The company also stands to collect a small fee each time a piece of data is recorded in the series.

The company provides the DGML format and the working version of the software, but not everything. Paoli said the company keeps some of its technology private, including AI models it has fine-tuned to read documents, and can sell those or other tools to businesses.

“Everyone’s business style is changing. And if there is a company you know that is not true, you need to tell me, because I haven’t met them yet,” said Paoli in an interview.

Docugami has raised nearly $13 million to date, including a $10 million seed round in 2020 that attracted the first investment in Grammarly’s history.

Partnership: Paoli met Patrick O’Meara, CEO of Inveniam, a few months ago, through a former Microsoft colleague who was one of O’Meara’s advisors. They soon realized that they were working on the same idea from different places.

Inveniam, founded in 2017, helps large investors track hard-to-value assets, such as office towers, private loans and infrastructure. It monitors the documents behind those assets and flags changes as they happen, and its clients include some of the largest private equity funds, according to O’Meara.

What was lacking was a consistent way to break those books into verifiable chunks. That’s what Docugami offers.

“We don’t put the data itself on the chain, only the fingerprint of the document. Change one bit, one byte, one pixel, and the hash won’t be the same,” O’Meara said.

The blockchain comes from Mantra, a crypto company owned by John Patrick Mullin. Inveniam invested $20 million in Mantra last year and has agreed to acquire it outright. Mantra’s OM token collapsed in April 2025, wiping out billions of dollars in value.

Paoli said the project uses the underlying blockchain, not a token.

“Crypto as an industry has undergone many changes in the last 18 to 24 months, and it is growing in many ways. This is a real use case with a fundamental value, not just pure speculation,” said Mantra’s Mullin in an interview.

The result is a division of labor: Docugami turns documents into data, Inveniam verifies it and delivers it to customers, and Mantra provides the chain where evidence is recorded.

The DGML specification, sample scripts and reference code are at dgml.io and GitHub.