Takuma Domoto · Tokyo, Japan
I build LLM systems for work where being wrong is not an option — and I sell them to the executives who sign off.
Five years selling and deploying AI and robotics into Japanese manufacturers and logistics operators. Now shipping production Claude systems into regulated professional work.
Automating a tax accountant's filing work — with the accountant still signing
A certified tax accountant firm in Tokyo. The work: year-end closing and statutory filing — reading receipts and ledgers, generating journal entries, building trial-balance comparisons, consumption-tax worksheets, and statutory breakdown schedules. Errors here are not bugs; they are professional liability.
I delivered four production systems as Claude Skills running inside the customer's Claude Cowork environment, and ran the engagement alone: discovery, requirements, implementation, delivery, and adoption support.
The pipeline that all four systems share.
Three decisions that made it deployable
- The model adjudicates; it does not calculate.
Transfer, arithmetic, and format application run deterministically. Claude resolves only what is genuinely ambiguous — e.g. mapping a receipt description to an account — against versioned master tables, so every judgment is reproducible and explainable. - Uncertainty is visible, per item.
Every output carries a three-level confidence flag (confirmed / review recommended / review required). Review effort tracks actual uncertainty instead of document count. - Delivery is gated on the books balancing.
Reconciliation — debits equal credits, net income matches the trial balance — is a hard output condition wired into the pipeline, not a report someone might read.
Operationally: the firm's own Excel templates are preserved (formula cells are detected and left untouched) so audit traceability survives; per-client differences live in CSV master tables; a new client is onboarded by deriving mappings from a single correct input/output pair.
Status: all four systems in production. In August 2026 a second regulated practice — a labor & social-security attorney firm — signed, showing the approach transfers.
A ¥1.0B automated warehouse, carried through four layers of decision-makers
At RENATUS ROBOTICS (AI × robotics automated storage), I lead solution sales and business development. I closed a ¥1.0B (~US$6.5M) deployment — the company's second contract ever — over a twelve-month cycle.
What moved it
- I owned the technical case personally. Analyzed months of the customer's inbound/outbound logistics data, modeled seasonal and intraday peaks, sized the equipment configuration, produced plan and elevation drawings, and built the ROI and payback model.
- I mapped the approval structure before selling. Who decides, and what each layer needs in order to defend the decision one level up. The same system, re-explained four times in four different languages: operations, finance, management, group governance.
- I stayed after the signature. Software, hardware, and maintenance requirements; 150+ pages of construction filings; municipal and developer approvals; installation, testing, acceptance.
How I think about enterprise AI
Adjudication over generation
In high-stakes workflows, the model's job is to resolve ambiguity inside a deterministic pipeline — not to produce the deliverable unchecked.
Failure must be visible
Confidence flags and reconciliation gates turn "trust the AI" into "check exactly here."
The approval chain is the real interface
An enterprise adopts what its people can defend internally. Design for the ringi, not only for the end user.
ringi (稟議) — the internal approval memo that carries a Japanese enterprise decision upward through each layer of management.