Enter individual clients' tax return documents
AI reads withholding slips, deduction certificates, and medical receipts from individual clients and enters them into tax software. Staff check unclear items.
- Runs on your own PCs and servers
- Use Azure OpenAI, Amazon Bedrock, Google Vertex AI, or an in-house model
- Its engine is Descarty's open-source Dagu (GPL-3.0)
The problem today
Individual clients send withholding slips, payment records, deduction certificates for life insurance premiums and the like, medical receipts, and receipts for hometown tax donations, on paper, as photos, or as PDFs. Between mid-February and the March 15 deadline, staff read each one, enter it into the tax software, and compare with last year to find anything missing.
The workflow in Kitewell
- Gather email attachments and files in shared folders by clientEmail
- AI identifies each document and reads the payer, amount, deduction type, and other itemsAI judgment
- AI totals medical receipts and lists documents from last year's return that haven't arrivedAI judgment
- Staff check unclear items and missing documentsApproval
- Open the client in the tax software, enter each document, and save the screensDesktop
- A tax accountant reviews the returnHuman task
What AI does
Reads the payer, amount, deduction type, and other items from documents in varied formats, and shows where in the document each value came from. Totals medical receipts by patient and provider, and lists documents on last year's return that haven't arrived this year.
Where people decide
Staff decide which deductions apply and what to ask the client. A tax accountant reviews the return.
Prove Kitewell on your own systems
Try it free for 14 days on your own PC. To see it working first, ask for a 30-minute demo. There is also a one-month onboarding (¥500,000 before tax), where we build your first job together.
What to check when you try this work
- How reliably each kind of document can be read, including handwritten receipts
- Which documents the Mynaportal link already provides, and which need reading
- Which AI model handles client information, and where it runs
How the onboarding runs
- Week 1Requirements and success criteria
- Weeks 2 to 3Build
- Week 4Verify on production screens
- Final dayReport
One-month onboarding
- Length
- One month
- Cost
- ¥500,000 (tax excluded). Your staff build, with the developer beside them
- If you go on to a license
- The onboarding fee is credited to the first year's contract
- Without the onboarding
- At Team scale, start on Team monthly (¥14,800 per computer a month, tax excluded), and cancel in any month
You can try it first on your own PC, in the free 14-day trial.
Frequently asked questions
- Doesn't the Mynaportal link make this unnecessary?
- The Mynaportal link pulls deduction certificates and other information through Japan's government portal and fills them into the return. It requires the client's My Number Card, and data is available only from issuers that support the link. Use it for what it provides; this workflow covers documents that arrive on paper or as PDFs.
- Do our documents and screens leave the company?
- Kitewell runs on your own PCs and servers, and Descarty never receives your business data. You choose the AI model: a cloud API, or an in-house or local model with an OpenAI-compatible API. With an in-house model, no data leaves the company. With a cloud model, what the AI reads is sent to the model's provider.
- What if the AI misreads something?
- A person checks what the AI produced before the work goes on. In a sheet, every value the AI reads shows the quote it came from, and rows below your confidence threshold go to a person. Reviewers can send work back with notes on what to fix, and who approved what is kept in the run history.
- What happens when a screen changes or a run stops?
- When a screen changes, the AI reads it again, relearns the step, and carries on. The relearned step is saved, and later runs repeat it. If a run still stops, ask the assistant right there, and it finds the cause and proposes a fix. A proposal is not saved until a person applies it.
- What happens to the RPA and systems we use today?
- Keep them. Kitewell operates screens the way a person does, so it works with systems that have no API. Move the robots that keep stopping and the work you gave up automating because it needs judgment, and leave fast, fixed processing in your current RPA.
- What does it cost?
- ¥14,800 per computer per month, or ¥148,000 per computer when paid yearly, before tax. A computer costs the same whether it builds or runs. You can try it free for 14 days. The one-month onboarding, where we build your first job together, is ¥500,000 before tax, credited to the first year if you go on.
Related use cases
Collect clients' monthly documents and remind them
Track each client's documents from incoming email. For clients still missing something, AI drafts a reminder naming it, and staff approve it.
Transfer year-end figures into tax software
AI reads each client's trial balance, maps the accounts to your tax software, and enters the amounts. Staff check unclear mappings and mismatched totals.
Check and enter year-end tax adjustment forms
AI reads employees' year-end adjustment forms and certificates, rechecks the deductions, and enters them into payroll software. Staff check differences.
See this work in a demo
A 30-minute demo shows this workflow running. The 14-day free trial lets you check it on your own systems.