Hebbia
| Company type | Private |
|---|---|
| Industry | Artificial intelligence, Automation, Software |
| Founded | August 2020 |
| Founder | George Sivulka |
| Headquarters | New York City , United States |
Key people | George Sivulka (Founder) |
| Products | Matrix (AI document analysis platform) |
| Website | hebbia |
Hebbia is a US-based technology company focused on developing artificial intelligence and automation solutions tailored for financial and legal research sectors. Established in 2020 by George Sivulka, who was pursuing a PhD at Stanford University at the time, the firm operates out of its headquarters in New York City.
The company has attracted investment from leading venture capital firms such as Andreessen Horowitz, Index Ventures, and Google Ventures (GV), as well as prominent individual investors including Peter Thiel, Eric Schmidt, and Jerry Yang.
History
Hebbia was founded in August 2020 by George Sivulka during his doctoral studies at Stanford. Its first offering was a semantic search engine that utilized large language models to improve search functionality within documents.
In 2022, Hebbia introduced Matrix, a platform leveraging natural language processing to extract information from various document types—such as contracts, presentations, spreadsheets, transcripts, and regulatory filings—serving industries like finance, law, and other knowledge-intensive fields. By 2025, Hebbia integrated large language models from OpenAI into the Matrix platform to further boost its performance.
Product and technology
Matrix, Hebbia’s main product, is a software platform designed to analyze documents including PDFs, spreadsheets, and slide decks. Users can pose questions in everyday language, and Matrix responds with answers accompanied by linked source references.
Within the financial sector, Matrix aids asset managers, investment banks, and private equity firms by streamlining due diligence processes related to mergers and acquisitions as well as investment analysis. It efficiently processes extensive collections of documents such as virtual data rooms, contracts, market intelligence reports, and compliance filings.
Legal professionals employ Matrix to support transactional and litigation processes by identifying key contractual clauses, conducting M&A due diligence, and performing document comparisons.
Funding
Since its inception, Hebbia has secured over $160 million through venture capital funding. Initial investment rounds in 2020 included backing from Peter Thiel and Floodgate. In 2022, the company raised $30 million during a Series A round led by Index Ventures.
In 2024, Hebbia completed a $130 million Series B financing round spearheaded by Andreessen Horowitz with participation from Index Ventures, GV (Google Ventures), and Peter Thiel. Distinguished individual investors such as former Google CEO Eric Schmidt and Yahoo co-founder Jerry Yang also contributed.
Acquisitions
FlashDocs Acquisition (2025): In 2025, Hebbia expanded its capabilities by acquiring FlashDocs, a startup launched in 2024 by Morten Bruun and Adam Khakhar that specialized in generative AI for creating automated slide decks. FlashDocs generated thousands of client-ready presentation slides daily based on structured inputs. This acquisition broadened Hebbia’s platform scope from document retrieval and agentic workflows to comprehensive artifact generation—automating investment memos, diligence reports, and board presentations—and advancing end-to-end AI workflow automation within financial services.
Research
In 2025, Hebbia researchers Jake Skinner and Davis Li released Who Evaluates the Evaluator: Reaching Autonomous Consensus on Agentic Outputs, presenting a consensus-driven methodology for assessing large language models (LLMs). Their technique combined permutation-based statistical analysis with comparisons across multiple models to produce more dependable performance evaluations. Additionally, they developed the Financial AI Benchmark, a tool designed to assess model effectiveness across financial workflows. These contributions underpin Hebbia’s model orchestration framework and highlight the importance of rigorous multi-model evaluation in enterprise AI applications.