RootCause.ai (operating as Perceptura, with offices in San Francisco and London) is a causal decision intelligence platform that lets enterprises simulate choices before they make them. It builds a dynamic, continuously updating digital twin of a business’ operation. This establishes a verifiable model of how things actually work, and uses it to answer the question every leader really cares about: not what happened, but why, and what happens if I act.
Most enterprise analytics stop at correlation. Dashboards and machine-learning models are excellent at telling you two things moved together, but not whether one caused the other. This is the crucial missing piece that is needed before changing a price, a policy, or a maintenance schedule. Rigorous causal analysis has historically been the fix, but it was slow, bespoke, and expensive: the preservation of a handful of the largest companies with armies of specialists and months to spare on a single question. RootCause.ai turns that capability into a repeatable, scalable platform. It connects to data from any system, spreadsheet, or database in its original format, automatically discovers the cause-and-effect relationships shaping performance, builds a digital twin of the relevant business function, and lets teams run simulations before committing to a decision. Work that once took months of data engineering and manual analysis happens in minutes or hours.
Key credentials:
RootCause's founding team pairs rare technical depth in causal AI and generalized machine learning with a proven record of scaling enterprise software from zero to real revenue.
Dr. Ayman Elhalwagy — Co-Founder & CEO. A PhD in Computer Engineering and dataset-fusion and generalized-ML expert whose research has been deployed at NASA and UKAEA. He previously founded and scaled an AI consultancy before starting RootCause, and now drives the company's product and research vision.
Dr. Jonas Skackauskas — Co-Founder & CTO. PhD in Artificial Intelligence and an expert in combinatorial optimization, with research deployed at Intel. He leads RootCause's technical architecture, cloud infrastructure, and security.
Jake Friedenberg — Co-Founder & COO. A repeat go-to-market leader with a track record scaling revenue from zero to nine figures. Early go-to-market hire (employee #3) at Tonic.ai and previously at npm. Now leads RootCause's operations, GTM, and sales.
Picks and shovels for the AI decision layer. RootCause sits deliberately between data infrastructure (Snowflake, Databricks) and the BI/visualization tools built on top of it (Tableau, PowerBI). As enterprises shift analytics budgets from static dashboards toward AI-native decision intelligence, that position makes RootCause a natural control point rather than a competitor to the platforms it plugs into.
Solving the industry's hardest technical problem. Causal discovery at enterprise scale has been a computationally intractable problem for decades, traditionally requiring PhD teams and expensive controlled experiments. RootCause.ai productizes it - automated, generalized, and continuous - collapsing months of expert work into minutes. That's the classic shape of a category-defining platform: take something scarce and expensive and make it standard.
Right team, right moment. Repeat AI founders combined with operators who've scaled real enterprise software businesses is a rare pairing. The team's NASA, Intel, Caterpillar, and npm pedigree gives them both the technical depth to build something genuinely novel and the go-to-market experience to sell it into large organizations, timely as AI-governance expectations are shifting from "can it predict?" to "can it explain?"
Enterprise pulls ahead of the round. The team is already in front of operations, finance, and commercial leaders at global logistics, financial services, and telecom companies, with quantified impact: 95%+ analysis-effort reduction, 3x forecast accuracy, and per-project value stretching into the tens of millions. That's tangible ROI in the language enterprise buyers actually budget against.
How it works, simply: Ordinary business analytics tells you that two things happened at the same time - sales dipped when a campaign ran, downtime rose when a supplier changed. It can't reliably tell you which one caused the other, so acting on it is guesswork. RootCause.ai reads a business's own data, works out the genuine cause-and-effect relationships behind performance, and assembles them into a digital twin of how the operation works. You can then test a decision on the twin, like a pilot in a flight simulator, and see the likely outcome before doing it for real.
The problem it solves: Enterprises make thousands of consequential decisions on correlation and gut feel because true causal analysis was too slow, too costly, and too specialized to use routinely. A single root-cause investigation could take skilled analysts days per metric, untenable across hundreds of KPIs in dozens of countries. That gap means wasted spend, missed risks, and slow reactions. RootCause.ai removes the cost, time, and expertise barriers so causal rigor can be applied to everyday operations, not just rare strategic bets.
Key features and benefits:
