Why I Co-Founded CPAAutomation.ai
Giving professionals more time for the work that requires their judgment.
By Ray Sang, CPA, CISA, CITP — Co-founder, CPAAutomation.ai
I helped build CPAAutomation.ai because I believe the time of an accountant, finance professional, or lawyer should be spent understanding problems and making sound decisions—not repeatedly moving information between documents and systems.
Consider a routine assignment. Before a professional can reach a conclusion, someone must collect the supporting documents, identify the relevant information, organize the data, prepare the analysis, and assemble the work for review. Those steps matter. But they should not consume so much attention that little remains for the questions that require experience and judgment.
My ambition is to make that work easier without weakening the standards behind it. I want professionals to have practical tools that help them complete an assignment, understand the result, and remain responsible for the decisions they make.
That is the purpose behind my work at CPAAutomation.ai.
What My Experience Taught Me
My background includes audit and assurance at Deloitte, finance transformation at SentinelOne, and finance systems leadership at Plaid. These roles have given me responsibility for both the work itself and the systems and controls that support it. Board Resume Board Resume
I approach automation as both an accountant and a software builder. The accounting question is whether the result is complete, accurate, and supportable. The engineering question is how to make the process repeatable without requiring someone to reconstruct it every time.
At Google Cloud, for example, I built automation that reduced contract identification from three days to real time. That was a specific process improvement, not the elimination of contract review or professional judgment. The distinction matters: automation can remove a substantial amount of preparation while leaving the important decisions with the people qualified to make them. Board Resume
That experience informs how I evaluate software. I do not begin by asking how many AI features a product contains. I ask what work it removes, what evidence it preserves, and what the reviewer still needs to do.
A faster process is valuable. A process that people can understand, review, and use consistently is more valuable.
Why We Are Building a Connected Platform
CPAAutomation’s original story began with Ian Stewart watching his mother, a CPA, spend long hours on paperwork. His interest in software led to a document-analysis project that developed into a broader platform for accounting, finance, and legal professionals. CPAAutomation
My experience brought me to the same problem from another direction: how to turn professional knowledge into working systems.
For me, extracting information from a document is a useful beginning, but it is not the whole assignment. The information may need to be checked, compared with another source, incorporated into an analysis, explained in a memorandum, and delivered for approval. I want our software to support that larger sequence of work.
That is reflected in the tools we are building. Universal Document Analysis helps convert documents into structured data that users can review and export. Form Fill uses source information to complete recurring forms. Inkwise supports professional writing using selected reference materials and citations that reviewers can follow back to the sources. CPAAutomation
The AI Analytics Suite extends that approach to accounting work, including transaction reconciliation, variance analysis, and supporting schedules. The platform also includes tools for client evidence requests, engagement coordination, and electronic signatures. CPAAutomation
These products address different tasks, but my reason for bringing them together is the same. I want less effort spent managing the movement of work and more attention given to its substance.
The goal is not simply to add another application to a firm’s existing collection. It is to make a complete assignment easier to manage.
Professional Judgment Must Remain Central
My legal education at Fordham adds another perspective to my accounting and engineering background. It reinforces the importance of distinguishing an assertion from a supported conclusion. Board Resume
That distinction is central to how I think about AI.
A polished memorandum is not enough. What supports the conclusion? A completed reconciliation is not enough. Which items remain unresolved? A populated form is not enough. Does the information belong in those fields, and has someone checked it?
I want automation to make those questions easier to answer—not easier to overlook.
For that reason, I see professional review as part of the product, not an inconvenience to remove. Our AccountingClaw and LegalClaw tools are designed to prepare work for professional review, with final judgment and approval remaining with the accountant or legal professional. CPAAutomation
The same principle should guide access permissions, confidential information, and the boundaries placed on automated actions. Before a system acts, the team should understand what it may do, when it must stop, and who is responsible for reviewing the result.
My standard is straightforward: automation should reduce the effort required to produce sound work, not reduce the responsibility for that work.
Making Automation Practical for More Firms
My ambition extends beyond building tools for organizations with large technology budgets.
I want a small accounting practice, an independent professional, or a growing finance team to benefit from automation without first having to establish an internal software department. A team should be able to start with one recurring problem, evaluate the results, and expand from there.
I also recognize that a standard product will not fit every process. A useful solution may require changes to the workflow, connections to existing systems, or controls designed for a particular organization.
That is why CPAAutomation also offers hands-on consulting. Our team works with clients to define the problem, build the software, and support deployment, documentation, and training. The service includes custom workflows, document processing, AI assistants, and AI governance. CPAAutomation
For me, this is an important part of the mission. I do not want to stop at explaining what technology could do. I want to help people put it to work.
Building Capability, Not Dependence
Education is another reason I care about this work. Alongside building software, I have taught accounting and other professionals about technology adoption, including blockchain, fraud, cybersecurity, and business intelligence through the AICPA. Board Resume
I want professionals to understand enough about their tools to question the output, recognize limitations, and improve the process. They should not have to become software engineers. But they should remain informed participants in how their work is performed.
That is also my longer-term ambition for CPAAutomation.ai: to become one place where accounting and finance teams can find useful software, practical implementation support, and the knowledge to use both responsibly.
Success, to me, would mean an accountant has more time to investigate an unusual balance, a finance team has more time to explain what changed, and a professional can finish an engagement without spending the evening repeating work that a well-designed system could handle.
I helped build CPAAutomation.ai to make that outcome more attainable.
The purpose is not to make professional expertise less necessary. It is to give that expertise more room to matter.