Basis Attracts $100 Million From Investors in Series B Financing
The AI startup reshaping how accounting firms get work done.
Basis, the artificial intelligence startup building autonomous agents for the accounting profession, has reached a $1.15 billion valuation after closing a $100 million Series B financing round, vaulting to unicorn status less than three years after emerging from stealth. The speed of the company’s ascent reflects a broader investor conviction that AI-driven automation in professional services is no longer speculative but is instead producing measurable returns at the industry’s most prominent firms. Accel led the round, with participation from GV, formerly known as Google Ventures, and former Goldman Sachs Group Inc. Chief Executive Officer Lloyd Blankfein. Existing backers including Khosla Ventures doubled down on their earlier commitments.
The financing positions Basis at the forefront of what has become one of the fastest-growing segments in enterprise technology. The AI-in-accounting market was valued at approximately $7.5 billion in 2025 and is projected to grow at a compound annual rate exceeding 44% through 2031, according to industry research estimates. The U.S. accounting services industry alone generates roughly $157 billion in annual revenue across more than 85,000 firms, many of which are contending with an acute and worsening talent shortage. The unemployment rate for accountants and auditors stood at just 2% in 2025, and an estimated 75% of certified public accountants in the current workforce are expected to retire within the next decade. These structural pressures have created urgent demand for technologies that can reliably automate high-stakes financial workflows, and Basis has moved aggressively to fill that gap.
The New York-based company, founded in 2023 by Matthew Harpe and Mitchell Troyanovsky, has built autonomous AI agents designed to handle the kind of labor-intensive, multi-step accounting work that has historically required large teams of trained professionals. Its technology ingests structured and unstructured data from client documents, accounting systems and enterprise resource planning platforms, then analyzes financial records, extracts relevant fields, performs calculations and applies validation rules within established compliance frameworks. Human accountants remain in the loop for final review and approval, but the agents handle the heavy lifting of data gathering, drafting and workflow progression. The company recently demonstrated what it described as the first AI agent to autonomously complete an end-to-end 1065 partnership tax return, a milestone that underscored the depth and ambition of its technical roadmap.
Basis is not operating in a vacuum. The AI accounting category has attracted a growing roster of competitors in recent years, including Botkeeper, which pairs automation with human bookkeeping oversight, Truewind, which targets venture-backed startups with AI-powered bookkeeping and financial modeling, and Numeric, which raised $28 million in 2024 to develop generative AI tools for close automation. Established players like Intuit and Xero have also integrated machine learning into their platforms to automate transaction categorization, reconciliation and invoicing. What distinguishes Basis, according to its investors, is the depth and autonomy of its agents. Rather than offering copilot-style assistance or automating narrow subtasks, the platform executes complex, multi-step workflows end to end across tax, audit and client accounting services. The company is now deployed at approximately 30 percent of the Top 25 accounting firms in the United States, with notable customers including Boulay PLLP, Clark Nuber PS, MarksNelson LLC, Pinion LLC and UHY LLC.
Miles Clements, a partner at Accel who led the investment, pointed to the company’s engineering rigor as a key differentiator. He noted that teams with strong foundational architecture tend to pull ahead quickly in emerging AI categories, and that Basis is well positioned to define the accounting AI market as it matures. Vinod Khosla of Khosla Ventures said the platform is already driving 20 to 50% efficiency gains across practices at leading firms and predicted it would deliver the same transformative impact to accounting in 2026 that AI coding tools brought to software engineering in 2025. Prashant Mital, Applied AI Lead at OpenAI, praised the company’s technical accomplishments, describing it as working at the frontier of production-grade, long-horizon agent development and noting that the collaboration between the two organizations has helped shape the broader future of AI agents.
The Series B brings total equity raised by Basis to roughly $134 million, following a $34 million Series A closed in December 2024 that was led by Khosla Ventures and attracted backing from NFDG, the investment fund founded by former GitHub CEO Nat Friedman and former Apple executive Daniel Gross. That round also drew participation from OpenAI board members Larry Summers and Adam D’Angelo, as well as Google’s chief scientist Jeff Dean. The latest financing added new names to an already notable cap table, with individual investors including Replit CEO Amjad Masad, Hugging Face CEO Clem Delangue, Box CEO Aaron Levie and Claire Hughes Johnson, the former Stripe executive. The breadth of the investor base reflects a growing consensus among technology leaders that AI-driven professional services represent one of the most significant near-term market opportunities in enterprise software.
Matt Harpe, the company’s CEO and co-founder, said the capital will be used to accelerate platform development, with a particular emphasis on building more advanced agents capable of handling increasingly complex workflows across tax, audit and advisory services. Basis also plans to expand its engineering and machine learning teams, focusing on talent that can improve model reliability, workflow depth and enterprise integrations. Harpe, a former Boston Consulting Group and SoftBank executive, co-founded the company alongside Troyanovsky and a team of researchers from Columbia University, MIT and Harvard, bringing together operational experience and deep technical expertise in probabilistic programming and machine learning. The company has also adopted what it calls an “agent-native” organizational structure internally, with a dedicated team called Atlas responsible for building AI agents across its own engineering, sales and talent operations.
The broader significance of the round extends beyond a single company’s balance sheet. While much of the venture capital flowing into artificial intelligence has gone to foundational model developers and horizontal infrastructure plays, the rapid growth of Basis suggests a parallel and potentially larger opportunity in building domain-specific agents that solve narrowly defined but economically significant problems in regulated industries. Accounting, with its combination of high complexity, rigid compliance requirements and chronic labor shortages, has emerged as one of the clearest proving grounds for this thesis. With a client roster that includes some of the most prominent names in American accounting and a valuation that now exceeds $1 billion, Basis has staked its claim as the category leader in agentic enterprise AI for professional services. For the industry’s incumbents, the question is no longer whether autonomous agents will reshape the profession, but how quickly.
