Fundamental exits stealth as $1.2 billion unicorn, armed with $255 million for tabular AI
Co-founded by Israeli entrepreneur Gabriel Suissa, the startup counts Wiz CEO Assaf Rappaport and Tel Aviv VC Hetz among backers, with part of its R&D rooted in Israel.
For all the excitement around generative AI, most enterprise decisions are still made from rows and columns rather than paragraphs and pixels. Fundamental, a startup with Israeli leadership and a Tel Aviv footprint, is betting that this neglected world of tabular data represents AI’s next frontier. The company said it has raised $255 million at a $1.2 billion valuation to commercialize a model designed to predict outcomes directly from corporate databases.
The funding, split between a $30 million Seed and a $225 million Series A, was led by Oak HC/FT with participation from Valor Equity Partners, Battery Ventures, Salesforce Ventures, and Israel’s Hetz Ventures. High-profile angels include Israeli Assaf Rappaport, co-founder and CEO of Wiz, alongside Perplexity CEO Aravind Srinivas, Brex co-founder Henrique Dubugras, and Datadog CEO Olivier Pomel.
Fundamental is making a contrarian bet that the real economic prize lies in the billions of spreadsheets and database tables that dictate pricing, risk, logistics, and medical decisions. “NEXUS is the OS for business decisions,” said CEO Jeremy Fraenkel, who co-founded the company with fellow DeepMind alumni.
Related articles:
- IBM invests in Israeli startup Anima to bring “vibe coding” into the enterprise
- “Israeli startups are extremely tethered to the US market, and we don’t see a way to build a big company from Israel without relying on the US market”
- “A reckoning is coming. We are currently sitting in a public market bubble that inevitably spills into the private markets”
While headquartered in the U.S., the company carries a distinctly Israeli imprint. Co-founder Gabriel Suissa is Israeli, and the investor roster features prominent figures from the country’s tech ecosystem, most notably Rappaport. Fundamental also maintains R&D activity in Israel, employing engineers who previously worked at AI21 Labs.
Fundamental argues that its model, NEXUS, was trained on billions of tabular datasets using Amazon SageMaker HyperPod and is designed to ingest raw tables with minimal configuration, often “a single line of code,” the company says. Once connected, it automatically learns dependencies across rows and columns, producing predictions for use cases such as price optimization, energy demand, financial fraud, and hospital readmissions.
The startup has already signed seven-figure contracts with Fortune 100 companies and announced a strategic partnership with Amazon Web Services, allowing customers to purchase and deploy NEXUS directly through their AWS dashboards.
