Sapien Raises $8.7M Seed Led by General Catalyst

This funding from top investors in AI and finance will support larger-scale research efforts to build the AI that helps humans run companies.

At Sapien, we're building autonomous coworkers for financial analysis at enterprises that are already turning error-prone, days-long processes into precise, insight-rich analysis in minutes. We’re excited to announce our $8.7M seed fundraise led by General Catalyst, with significant participation from Neo. We’re also incredibly grateful to have the support of an amazing group of angel investors including founders, executives, and top people from Ramp, Cognition, Google, OpenAI, Stripe, Notion, Adobe, Pilot, and more. See the Sapien founding team below (left to right): Arya Grayeli (Chief Scientist), Pranav Ravella (CTO), Ron Nachum (CEO).

"The role of CFOs has changed over the years, and now puts the entire CFO stack at an inflection point that’s ripe for disruption. In our view, Sapien is productizing the cutting edge of AI research for CFOs, with the opportunity to truly revolutionize how finance teams operate"

Ken Chenault, Chairman and Managing Director of General Catalyst, and the former Chairman and CEO of American Express

The “traditional CFO” is dead.

For decades, CFOs have spent their time on rearward-looking number crunching to track and approve company budgets. This used to be purely a defensive role, but today, CFOs play on both sides of the ball. They are quickly evolving to be a company’s most dynamic offensive weapon—driving investments into growth, creating operational efficiencies, and unlocking new opportunities for scale.

These new-era CFOs need to understand every facet of their companies in order to make these critical decisions, especially when it involves capital allocation. This is not the reality at today’s enterprises.

The world’s biggest companies are flying blind.

The financial infrastructure of the world’s biggest companies is scarily fragile.

Many Fortune 500 giants manage billions in revenue with unwieldy, multi-million row Excel spreadsheets. Global enterprises grapple with 136 disjointed, outdated ERPs that would require years and millions of dollars to reconcile. These same companies even spend years relying on models that make eight-figure errors. These mistakes persist because the underlying models are incredibly complex, hard to verify, and often require institutional knowledge that is lost as their creators leave the company.

These are just a few examples of what you’d see in an enterprise finance team. It’s an open secret that the finance function is plagued by critical, easily avoidable mistakes every single day.

And the reason is simple: humans are doing work that should be left to machines. We’re trying to analyze massive Excel spreadsheets and wrangle information from antiquated ERPs and CRMs (the amount of companies who actually don’t know how to get data out of these systems is frighteningly high).

The sheer volume and fragmentation of key information creates an error-prone process, forcing finance teams to rely on incomplete data for their most critical decisions.

This is the kind of work computers were built for—so why aren’t they doing it?

The industry is trying to fix a glaring gap with bandaids and baby steps.

Well—many are trying. This analytical gap is a critical daily challenge for CFOs, and while various solutions exist, they fall into a few key categories that miss the mark:

The Point Solution Patchwork: Tackle every problem individually and end up with 15 different SaaS solutions that (somewhat) solve each piece of the puzzle but also don’t communicate with each other. Now you’re spending more time managing IT and switching apps than you are analyzing data.

The Dashboard Delusion: Opt for a business intelligence solution that provides you with dashboards, colorful charts, and well-formatted tables. These look impressive, but they often don’t fit your specific company (many call this “BI without the I”). You have nice static visualizations, but you don’t have truly actionable, let alone flexible, insights.

The Chatbot Chasm: Jump on the AI train and opt for a solution that promises “AI.” This word has become overloaded, meaning anything from a spreadsheet autocomplete to a search tool. Not all AI is created equal. Many tools implement chatbots to fetch data on request, but they don’t actually understand it and they cannot meaningfully act on it.

The prevailing belief among CFOs has been to settle for these inadequate systems because "how could an automated system possibly understand MY company as well as I do?" And this skepticism has been well-founded. This was a problem that we didn’t have the technology to solve—until now.

The key is that enterprise finance isn’t just about numbers—it’s about the story behind each figure and how they connect across the entire organization. True financial intelligence requires a deep contextual understanding of every corner of a company, which is only just becoming possible at the forefront of AI research.

We’re taking the leap. Sapien deeply understands company financials, empowering teams to make correct decisions, faster.

We’re building the first AI-native system from the ground up that is designed to empower CFOs and the world’s largest finance teams.

By building the strongest underlying representation of company data—spanning siloed data sources from messy Excel sheets to outdated ERPs, clunky CRMs, and unstructured data pockets—we enable autonomous coworkers to do the work that humans shouldn’t. Companies onboard with Sapien by plugging into their ground truth data, enabling them to quickly ask the toughest questions they might throw at a human analyst. Instead of waiting days or weeks for insights, they now have an autonomous coworker that delivers them in minutes. Check out more about how we build our company representations here.

Sapien's deep understanding of each unique company gives it the flexibility to dynamically adapt and robustly tackle the most complex processes. For example, for manufacturing corporations, Sapien can analyze a year’s worth of raw transaction data across dozens of plants to uncover key mix impacts. For healthcare companies, Sapien can evaluate revenue and visit trends across hundreds of clinics to provide actionable growth recommendations. And for software companies, Sapien can compare unstructured data across large customer cohorts to better understand and harness growth trends. The examples are endless, and in each case, Sapien reduces over 100 hours of manual work to just five minutes of supervising the AI coworker.

Sapien propels finance teams to make better decisions faster, uncovering deeper insights than any analyst and has even caught multi-million dollar mistakes. We serve customers across manufacturing, services, and software businesses and have helped portfolio companies of top PE firms streamline operations. In the coming months, we will be expanding our deployment to a wider range of CFOs across all industries, scales, and geographies.

We prioritize the non-negotiables: reliability and transparency.

Answers are important. But in finance, answers are only worth as much as the assumptions they are based on. This is a huge part of why we’ve meticulously focused on pushing the envelope of what’s possible when it comes to agent observability and UX paradigms. Differing from most AI agent companies—we aim to build a deeply technical product for a purely non-technical audience. This limits how much underlying technical complexity can be made visible to users while also raising the bar for interaction quality.

A core part of our approach is how we build reliable agents, which you can read more about here. The key principle is using verifiable tools at the lowest levels, and this enables us to surface exactly how Sapien is breaking down and thinking about queries. And even further, the use of these tools significantly reduces hallucination risk. Finance teams can see exactly where Sapien is pulling data from and how it's transforming input information to produce outputs. 

Most centrally, users can challenge every assumption Sapien makes. Traceability and observability is one thing, but being able to intervene, course correct, and even educate the system is paramount. Challenges like “Think about the problem in this way” or course corrections like “Pull data from here instead” enable teams to not just get answers, but work together with Sapien. They can understand the process and key assumptions, while ultimately trusting and having the ability to build on top of the results produced by Sapien.

We’re building the instant & org-wide enterprise finance future.

We’re building a present where enterprise finance tasks that today take days or even weeks can be done in minutes. But importantly, our mission goes beyond simply enhancing human efficiency. We’re aiming to transform the very nature of human work and reimagine how companies run.

With Sapien, we’re uniquely positioned to democratize access to financial analysis in a way that has never been possible before. We’ve seen CEOs, marketing teams, PE operating partners, and even legal teams use or request access to Sapien. Why? Because they want financial insights without bogging down their analysts. And the power of this collaboration is further compounded by the new types of analyses we can unlock by marrying data across these teams.

What kinds of paradigm shifts can we see when finance teams can analyze 100 scenarios in an hour instead of 2 in a week? And what changes when these insights are deeper, autonomously identified, and guiding analysts to the most important points 100x faster than they’ve ever been able to. Humans will finally be able to focus on what they are best at: thinking strategically and driving innovation.

So this all begs our favorite question…

Can AI run a company?

Today, Sapien is the most advanced autonomous coworker for financial analysis at enterprises. But our customers deserve more, and our technical appetite to tackle impossible problems sets our sights on a bigger goal: building the shadow CFO.

As Sapien’s company representation improves and it learns more about every aspect of doing this financial analysis, we aim to abstract away the low-level analysis that humans shouldn’t be doing and create a deeply proactive system. Our goal is for Sapien to not just be a part of the process, but the irreplaceable layer between data and decision-making across all company scales that identifies multi-million dollar opportunities and risks while preparing CFOs to tackle them.

In the next 5 years, Sapien will arm every CFO with the newest information and insights about their company, preempting them with recommendations on what they should do next to drive growth. These CFOs should wake up in the morning knowing that their competitor had an earnings call, their sales team beat forecasts by 5%, and that 3 departments just uploaded new data into their ERP—and what they should actually do about this—all before their morning coffee. 

This is the future where humans are doing what they’re best at—innovating and thinking strategically—while AI does the grunt work that it takes to run companies.

We’re making it happen.

We’re incredibly excited to announce our $8.7M seed fundraise that brings amazing partners onto our cap table to realize this vision, including Niko Bonatsos & Max Rimpel from General Catalyst, Ali Partovi and the entire Neo team who have supported us since day 0, Bryan Baum at K5, and amazing angels including Russell Kaplen (Cognition), Claire Hughes Johnson (Stripe),  Sabrina Hahn (SH Fund), Scott Belsky (Behance & Adobe) and more.

With this fundraise, we’re excited to ship more powerful product iterations to customers, continue building the fastest-moving team of product-obsessed AI researchers out there, and push the limits of what's possible in this space.

We’re building a talent-dense, high-velocity team to tackle hard problems. If you’re deeply technical and excited about end-to-end ownership of impactful and product-oriented AI research, reach out here.

And of course, if you’re a CFO that wants to spend more time playing offense, request a demo and time to chat with us here.