JohnnyRebel

Fair, QuickBooks does market their software towards every small business owner not just accountants. But even so, small business owners still spend hours on entry and categorization. It's enough that most community colleges teach QuickBooks courses. We want to go further: let owners ask questions in plain language and get answers instantly, no menus or reports needed. Our upcoming versions will include agentic capabilities to send invoices, and do much more all from one AI UI.

We sidestep the “silicon psychologist” issue: the LLM simply interprets questions, while all numbers come from structured data. AI explains results, but it can’t rewrite the books.

Fair point—though to be clear, the LLM isn’t doing the math, just the interface. The numbers come from structured data, so accuracy isn’t left to chance. Where this really helps is for small business owners who are overwhelmed by QuickBooks data entry and classification. Our goal is to continually improve the experience, making bookkeeping as simple as possible.

I totally agree; the liability is real, which is why we don’t let the LLM “invent” numbers. We use the model as the interface, but all financial data comes from a structured database. In practice, it works like RAG: the LLM interprets the user’s question, retrieves the right data, and explains the result in plain English. That way the math is deterministic, the answers are grounded, and the AI layer just makes it accessible.

That’s exactly the question we’re testing. Will this feel like the future of bookkeeping, or just another tool? We’re launching the beta in a week and are eager to see how real users respond. Curious what you think would make this genuinely useful vs. gimmicky?

We thought about just building an AI plugin for QuickBooks, but decided to build our own platform instead. Harder path, but bigger upside.

We’re tackling ingestion primarily through direct bank connections. Users connect their bank and financial accounts, and transactions flow into our system automatically. From there, we store the data in a structured database and normalize it into a consistent internal format so it can be aggregated and queried reliably.

Right now, the ingestion layer handles most of the heavy lifting—parsing the raw feeds, mapping fields into a standard schema, and ensuring consistency across institutions. Our next version will include layering AI to help with classification and enrichment (e.g. categorizing ambiguous transactions, detecting anomalies, and filling in context where the raw data is thin).

So it’s a mix: the ingestion pipeline makes the data uniform, while AI helps make it more useful and accurate for analysis. As we move toward our “agentic” roadmap, we see AI playing a bigger role in automating the messy parts of ingestion as well.

The story of a bootstrapped AI-first bookkeeping app that lets small business owners talk to their financial data instead of wrestling with spreadsheets. Beta launching this September. Curious if HN thinks this is the future of accounting or just another shiny tool.