Rillet Hits $1 Billion: How a $100 Million AI Bet Is Rewriting the Future of ERP and Accounting

Michal Kosinski
14 hours ago
9 min read

Rillet has emerged as one of the clearest examples of how artificial intelligence is beginning to reshape enterprise software from the ground up. The AI-native accounting and enterprise resource planning startup has raised $100 million in Series C funding at a $1 billion valuation, reaching unicorn status only about two years after emerging from stealth.

The significance of Rillet’s rise extends well beyond another large venture round. The company is challenging a foundational assumption of enterprise software: that financial systems should primarily record transactions for humans to review and interpret. Rillet is pursuing a different architecture, one in which AI agents operate directly within financial infrastructure while humans retain control over judgment, governance and approval.

The result is a potentially important transition from ERP systems as passive systems of record to finance platforms capable of becoming active operating systems for businesses.

Rillet’s Rapid Rise to Unicorn Status

Rillet’s fundraising trajectory illustrates the extraordinary investor interest surrounding AI-native enterprise applications. The company has raised more than $200 million since emerging from stealth, with investors including ICONIQ, Sequoia, Andreessen Horowitz, Bain Capital Ventures, Oak HC/FT, Battery Ventures, FirstMark, Scale Venture Partners and Creandum.

Its latest $100 million Series C was led by ICONIQ, with participation from existing and new investors. ICONIQ general partner Seth Pierrepont joined Rillet’s board following the financing.

The company now has more than 600 customers, ranging from technology companies and rapidly growing AI businesses to organizations in industries such as healthcare, biotechnology, financial technology, logistics, professional services, waste recycling and entertainment.

Rillet’s growth is particularly notable because its customers are not merely experimenting with AI accounting capabilities. According to the company, organizations are replacing established financial systems with its platform.

Its customer acquisition mix provides another indication of the disruption underway:

Legacy platform source	Share of Rillet customers
Intuit	50%
NetSuite and Sage Intacct	30%
Oracle, SAP, Workday and Microsoft products	20%

The company also reported that new annual recurring revenue doubled during the three months preceding its latest financing.

That combination of customer growth, accelerating revenue and displacement of established enterprise platforms helped turn Rillet into an unusually compelling investment opportunity.

Why AI Is Challenging Traditional ERP Systems

Enterprise resource planning software has traditionally been designed around a relatively straightforward philosophy: collect, organize and preserve financial information.

The general ledger is the authoritative record. Employees enter transactions, accountants reconcile them, finance teams analyze the results and executives use reports to make decisions.

That architecture worked because humans performed most of the cognitive and operational work surrounding financial data.

AI agents change the equation.

An increasingly capable agent can interpret information, execute multistep workflows, retrieve relevant records, identify anomalies and perform repetitive accounting activities. But these capabilities become significantly more useful when the AI has direct access to structured financial infrastructure rather than operating as an external assistant disconnected from the accounting system.

This is the architectural argument behind Rillet.

Instead of placing an AI layer on top of an existing ERP, Rillet has designed its platform around AI agents from the beginning. Structured financial information flows into a real-time general ledger, while agents can operate within that environment and access the context necessary to perform financial work.

The distinction is important.

A conventional ERP can tell a company what happened. An agentic finance platform aims to help determine what needs to happen next.

From System of Record to System of Action

The most consequential idea behind Rillet is the transformation of the ERP from a passive repository into an active financial operating layer.

Traditional ERP software generally separates data storage from the work performed using that data. Spreadsheets, email, reporting systems and specialized applications often sit alongside the central financial system.

That fragmentation creates friction.

A finance professional may need to extract information from several systems, reconcile discrepancies, perform calculations, prepare reports and then return the resulting information to the ERP.

An AI-native architecture can potentially collapse many of those steps.

Rillet’s approach places AI agents inside the financial environment itself. Humans and agents can therefore work with the same underlying information, accounting rules and controls.

This creates several potential advantages:

Continuous financial operations: Agents can execute workflows without being restricted to conventional office hours.
Greater process automation: Repetitive accounting activities can be delegated to software.
Context-aware execution: Agents can work with structured financial information rather than isolated prompts.
Improved traceability: Actions can be recorded as part of an auditable workflow.
Human oversight: Finance professionals remain responsible for approvals and higher-value decisions.
Real-time information: Financial teams can potentially work from continuously updated records rather than periodic reporting cycles.

The broader implication is that enterprise AI may not simply automate individual tasks. It may change where those tasks are performed and how software itself is structured.

Agentic Finance Is More Than AI Bookkeeping

Calling Rillet an AI accounting company understates the ambition of the platform.

Accounting is the initial entry point, but the larger opportunity is the automation of the finance function.

Finance departments perform numerous interconnected activities, including bookkeeping, reconciliation, reporting, forecasting, compliance, controls, financial analysis and decision support. Automating one isolated task creates limited value if employees still have to manually coordinate the surrounding workflow.

Agentic finance seeks to connect those activities.

An AI agent could, for example, retrieve financial information, apply defined accounting policies, perform a reconciliation, identify an exception and prepare the relevant output for human approval. The value comes not simply from completing one task faster, but from reducing the number of manual transitions between systems and people.

That is why the concept has attracted significant attention from investors.

Sequoia investor Julien Bek described accounting as Rillet’s initial entry point while framing the broader opportunity as the reinvention of the finance function. The underlying thesis is that finance could become one of the major enterprise software categories transformed by agentic AI.

The Human Accountant Is Not Disappearing

The emergence of AI agents inevitably raises a difficult question: will accounting automation eliminate finance jobs?

Rillet’s position is that AI should augment finance professionals rather than simply replace them.

That argument is reinforced by the structural shortage of accounting talent in the United States. The accounting profession has experienced a decline in the number of students pursuing accounting degrees over an extended period, while organizations continue to require financial expertise.

A Controllers Council report cited in the supplied research found that 61% of finance leaders had experienced difficulty finding finance, accounting and CPA talent during the previous year.

At the same time, the U.S. Bureau of Labor Statistics has projected growth in accounting and auditing employment through 2034, suggesting that automation is not necessarily synonymous with declining demand for financial professionals.

The more plausible transformation is a shift in the composition of accounting work.

Routine data entry and repetitive reconciliation can increasingly become software functions. Human professionals can spend more time on analysis, judgment, controls, strategic planning and advising business leaders.

This distinction matters because accounting is ultimately not only about processing numbers. It is about understanding what those numbers mean and deciding how an organization should respond.

Governance Becomes Critical as AI Agents Gain Autonomy

Greater AI autonomy creates a corresponding governance problem.

When an AI system produces a recommendation, a human can evaluate the output. When an AI agent executes a sequence of financial actions, organizations need to understand exactly what the system did, which information it used and why it reached a particular conclusion.

Rillet has responded by developing governance capabilities that allow accountants to inspect and audit agent decisions, including the numbers used during calculations.

This becomes increasingly important as AI agents move from simple automation toward long-running, multistep workflows.

The more autonomous the system becomes, the more important auditability becomes.

Financial software must therefore balance three competing objectives:

Objective	Requirement
Automation	Agents should execute useful work efficiently
Control	Humans must retain appropriate authority
Transparency	Every important action must remain understandable and auditable

For public companies and highly regulated organizations, this balance is particularly important. Existing requirements can require human approval for transactions performed by AI systems, creating a governance framework that will likely evolve as agentic finance matures.

Security and Data Ownership Are Strategic Requirements

Financial systems contain some of an organization's most sensitive information. Customer records, payroll information, transactions, revenue data and strategic financial information cannot be treated like ordinary application data.

Rillet has therefore emphasized data isolation and model governance.

Its architecture includes model routing, allowing customers to direct AI requests toward foundational models from providers such as OpenAI or Anthropic. The company also says its infrastructure prevents those models from training on customer data.

Another important principle is the separation of customer information. Data from one organization should not become available to another organization through model training or shared contextual systems.

These controls illustrate an important principle for enterprise AI: model capability alone is insufficient.

A powerful AI agent becomes commercially useful only when organizations trust the infrastructure surrounding it.

The EY Alliance and Accounting Firm Ecosystem

Rillet is also attempting to establish credibility within the accounting profession rather than positioning itself exclusively as a software disruptor.

The company launched an alliance with EY for AI-native finance transformation and has reported partnerships with more than half of the Accounting Today top 20 CPA firms.

That strategy could become strategically important.

Accounting firms possess domain expertise, established client relationships and deep knowledge of financial controls. AI software companies, meanwhile, bring new technical capabilities.

Combining those strengths could accelerate adoption while helping organizations understand how agentic systems should be introduced into existing financial governance frameworks.

The partnership model also demonstrates that disruption does not necessarily mean eliminating incumbents. In some cases, new AI infrastructure may become a tool through which established professional services organizations transform their own operations.

What Rillet’s Funding Says About Enterprise AI

Rillet’s $1 billion valuation reflects more than enthusiasm for another AI application.

It signals growing investor confidence that enterprise AI will increasingly replace or restructure traditional software architectures.

The first generation of enterprise AI largely focused on adding conversational interfaces to existing products. The next phase is more ambitious: rebuilding core applications so that AI agents can directly participate in operational workflows.

This distinction may determine which companies become durable AI businesses.

Adding an AI assistant to legacy software can improve user productivity. Rebuilding the underlying system around AI can change the economics and workflow of an entire category.

Rillet is betting that finance and ERP represent one of those categories.

The Future of AI-Native ERP

If Rillet’s model succeeds, the ERP of the future may look fundamentally different from the systems businesses have relied on for decades.

Finance teams could operate continuously rather than waiting for periodic reporting cycles. Agents could monitor financial activity, identify exceptions and perform routine processes while humans focus on decisions requiring judgment.

The strongest systems will likely combine several characteristics:

Real-time financial data
AI agents capable of multistep execution
Deterministic accounting controls
Complete audit trails
Human approval mechanisms
Strong data isolation
Interoperability with existing enterprise systems
Domain-specific financial intelligence

The challenge will be ensuring that increased automation does not produce reduced accountability.

Enterprise finance cannot operate on the principle that an AI system is correct simply because it is sophisticated. Every consequential action must remain explainable, governed and reversible where appropriate.

Rillet Is Betting on the Operating System for Agentic Finance

Rillet’s rapid journey from stealth startup to $1 billion unicorn demonstrates how quickly AI is changing expectations for enterprise software.

Its $100 million Series C, more than $200 million in total funding, more than 600 customers and accelerating annual recurring revenue indicate significant market momentum. More importantly, its strategy reflects a broader architectural shift.

The next generation of enterprise software may not simply help humans use computers more efficiently. It may allow software agents to perform substantial portions of operational work directly inside the systems that businesses depend on.

Rillet is applying that idea to accounting and ERP, a category historically dominated by platforms such as Oracle, SAP, Workday and NetSuite. Its central proposition is that financial infrastructure should become intelligent, continuously active and designed for collaboration between humans and AI agents.

For finance professionals, the transformation could mean less time spent processing information and more time interpreting it. For companies, it could mean faster financial operations, greater automation and potentially smaller administrative burdens. For investors, it represents a much larger question: whether AI-native companies can genuinely replace some of the foundational software architectures built during the previous era of computing.

The answer will depend on execution, governance, security and customer trust.

But the direction is increasingly clear. AI is moving deeper into enterprise infrastructure, and companies such as Rillet are betting that the future of finance will not be an ERP system with an AI assistant attached. It will be an intelligent financial operating environment in which humans and autonomous agents work from the same financial truth.

For technology observers, including Dr. Shahid Masood and the expert team at 1950.ai, Rillet represents a useful case study in a much broader AI transition: the movement from software that records human activity toward systems capable of actively participating in business operations.