Automation and AI for accountants and finance teams
We build AI for accounting firms and in-house finance teams that handle the bookkeeping volume nobody wants to type: invoices, receipts, bank lines and client document requests. Accountants review what the software prepares, and everything lands in the packages you already use.
Where the work gets stuck.
Peaks with fewer people
Quarterly VAT returns, year-end work and annual reports all arrive at once, while qualified staff are hard to hire. Routine processing crowds out advisory work clients are willing to pay for.
Client documents arrive late and messy
Receipts, bank statements and contracts come in by email, WhatsApp photos and shared folders. Chasing missing items takes as much time as booking them.
Every client runs a different package
A single firm may work in Exact Online, Twinfield, SnelStart and AFAS across its client base. Standard tools cover part of the work and staff bridge the rest by hand.
How we improve accounting.
Each one is scoped around your systems and rules, and each one keeps a person in charge of the decisions that matter.
Invoice and receipt processing
Purchase invoices and receipts are read, matched to suppliers and given a suggested ledger account, VAT code and cost centre. Bookkeepers confirm or correct, and corrections improve the next suggestions.
Invoice approval agent
An agent checks invoices against purchase orders, contracts and your approval policy, and routes each one with the clause behind its recommendation. Budget holders approve and the agent books the result.
Bank reconciliation and period-end checks
Bank lines are matched to open items, and unusual postings, duplicate payments and VAT inconsistencies are listed before closing. The accountant works through the exceptions list and signs off the period.
Client document portal
Clients upload documents against a checklist per engagement, get automatic reminders and see what is still missing. Your team stops chasing paperwork by email.
Reporting across client administrations
Data from several accounting packages flows into one model for dashboards, benchmarks and signals such as liquidity dips. Advisers spot which clients need a conversation.
Tax and procedure search
Internal memos, tax guidance and firm procedures become searchable by question, with sources attached. Junior staff find answers faster and partners review less repeat work.
Built around the rules.
What we design for from the first week. Your legal and compliance people keep the final word.
GDPR (AVG) and client data
Accounting data includes salaries, BSNs and personal spending. We restrict access per engagement, keep processing in the EU and set up processing agreements your clients can rely on.
Wwft client due diligence
Accountants and tax advisers have Wwft duties. Tools that touch onboarding or unusual transactions keep records of checks and signals, and your staff decide what to report.
Tax retention duty
Dutch tax law requires businesses to keep their records for seven years. Source documents, bookings and approval trails are stored so they can be retrieved for the tax authorities.
E-invoicing and Peppol
EU plans under ViDA are moving B2B invoices towards structured e-invoicing. We build intake that handles Peppol UBL invoices alongside PDFs, so the flow keeps working as volumes shift.
Works with what you run.
If a system has an API, a database, an export or an inbox, we can build on it. These are the ones we meet most.
- Exact Online
- Twinfield
- AFAS Profit
- SnelStart
- Visma
- Yuki
- Nmbrs
- Silverfin
- Unit4
- Microsoft Dynamics 365 Business Central
Where to start.
Purchase invoice processing for one package
Invoice and receipt processing for the clients or entities that use one accounting package, with bookkeepers reviewing each suggestion. It gives measurable time savings quickly and a clear case for extending to other packages.
Talk it throughWhat it includes
- Intake from email, upload and Peppol
- Ledger, VAT and cost centre suggestions
- Review queue for bookkeepers
- Booking in Exact Online, Twinfield or AFAS
Related work.
- Finance operations
AI invoice approval agent that cites the policy clause behind every decision
100%of automatic decisions cite a policy clause- AI agents
- Human in the loop
- RAG
- Financial services
Open-source document parsing API and MCP server for financial documents
210+API endpoints- Document parsing
- OCR
- MCP server
Guides.
Build vs buy software: a decision guide for operations and product teams
A practical guide to choosing between off-the-shelf and custom software, with a cost comparison, a decision matrix and a checklist for your team.
10 min read
AI readiness checklist: what to answer before your first agent goes live
The questions to answer before your first AI agent handles real work, from process and data to approvals, monitoring and ownership.
10 min read
Further reading.
Document parsing with LLMs: extracting data from invoices and forms
How to turn invoices and forms into reliable structured data with LLMs, schemas, validation and a review queue for the cases that need a person.
4 min read
Human-in-the-loop AI agents: design patterns for production
The patterns we use to run AI agents safely in production: approval checkpoints, confidence thresholds, tool permissions, audit trails and fallbacks.
4 min read
Build vs buy software: when custom development pays off
A practical way to decide between off-the-shelf software and custom development, covering total cost of ownership, integrations and risk.
5 min read
Common questions.
Yes. All three have APIs we build against, as do most other Dutch accounting and payroll packages. Firms with a mixed client base can use one flow that books into each client's own package.
If your current tool covers your volume and packages well, keep it. A custom build pays off when you need approval logic, cross-package reporting or document types that standard tools handle poorly, and our build-versus-buy guide helps you weigh it.
Your bookkeepers and accountants. The software suggests and prepares, and a person confirms before anything is posted or filed. Every suggestion and correction is logged.
Access is scoped per client and per role, data is processed in the EU and models are used under terms that exclude training on your data. We can also run the whole setup inside your own cloud.