Unstructured: Making Data AI-Ready
Connecting enterprise data to large language models (LLMs)
Many academics, tech luminaries, and entrepreneurs believe that we are in the midst of an unprecedented artificial intelligence (AI) revolution, some arguing that the development of AI could rival the discovery of electricity. While history and hindsight will determine the scale of this technological leap forward, established tech giants and startups alike have embraced the technology.
Google has integrated AI directly into its search engine, possibly one of the company’s biggest changes to its browser in many years. Similarly, Microsoft announced it will directly integrate AI into the latest version of Windows. Underpinning both moves is the ability of organizations to seamlessly integrate data into large language models (LLMs) — AI models trained on a large corpus of proprietary data to uncover patterns in human language.
The crucial ability to package data for AI ingestion is exactly where AV portfolio company Unstructured has staked its claim. The company’s platform connects unordered, secure data to powerful LLMs through open-source libraries and APIs. Their ETL (Extract, Transform, Load) platform combines data from multiple sources into a large, secure, central repository that is AI-ready.
Packaging Data for Use With AI
Unstructured’s key innovation is its ability to write enterprise-grade connectors capable of capturing data wherever it is stored, transforming it into AI-friendly files for companies seeking to deeply integrate AI into their business. Unstructured is able to deliver data that’s curated, clean of artifacts, and most importantly, LLM-ready.
While AI adoption has rapidly grown across different organizations in one form or another, many enterprises in both the private and public sectors cannot leverage open-source AI due to privacy concerns. Unstructured has tapped into this early market demand by building a secure LLM-native ETL platform. While Unstructured has secured several key contracts with public sector agencies, the company has also piqued the interest of large private sector enterprises, with early adopters including major AI player Nvidia and consulting firm Booz Allen Hamilton.
What We Like About Unstructured
- Home
Rapid Revenue Acceleration
Unstructured has achieved meaningful total contract value in just under a year from public sector and enterprise-scale customers, each with significant expansion opportunities. Given the founder’s background contract execution and the roadmap to long-term revenue from government contracts will likely continue to accelerate. - Home
Defense Technology Overlay
Brian Raymond, the company’s CEO and an ex-CIA officer, brings expertise in defense technology, an area of increasing interest in the venture capital ecosystem. The early adoption by the defense sector strengthens the case for private sector sales, demonstrating that selling to government entities paves the way for corporate success. - Home
Harnessing the Momentum of AI
As user access to openly available AI and LLM modules such as ChatGPT becomes widespread, Unstructured's focus on top-secret grade security becomes even more attractive to large organizations.
How We’re Involved
Green D Ventures (for the Dartmouth community) sponsored our investment in Unstructured’s $25 million Series A led by Madrona Ventures. This opportunity was a strong match for Green D, as Unstructured Founder and CEO Brian Raymond earned his MBA from Dartmouth’s Tuck School of Business. Our investing teams regularly leverage their alumni connections to build relationships with entrepreneurs and VCs.
Other Alumni Ventures funds invested in the round include Arbor Street Ventures (for the Michigan community), Expo Ventures (for the USC community), and The Yard Ventures (for the Harvard community), as well as AV’s AI Fund and Total Access Fund. Other co-investors include Bain Capital Ventures, M12 Ventures, Mango Capital, MongoDB Ventures, and Shield Capital.
Want to learn more?
View all our available funds and secure data rooms, or schedule an intro call.
New to AV?
Sign up and access exclusive venture content.
Contact [email protected] for additional information. To see additional risk factors and investment considerations, visit av-funds.com/disclosures.