About
Agentic AI bookkeeping, built for how GCC firms actually work.
Nexus Ledger ingests messy bilingual voucher batches, invoices, statements, and salary records, turns them into auditable double-entry journal entries, routes them through the firm's approval chain, and produces a defensible trial balance — without sacrificing the controls that real auditors and regulators expect.
Who we serve
Two distinct buyers. Same workflow engine. Different access models.
Accounting firms
Bookkeeping practices, audit firms doing year-end work, outsourced finance providers, and CFO-as-a-service shops. They onboard one workspace, then run unlimited end-clients inside it — each with its own chart of accounts, periods, approval roles, and audit log. RLS keeps every client's books separate even from inside the same firm.
In-house finance teams
Group accounting, regional finance functions, and consolidated holding companies running multiple legal entities. They onboard one workspace per group, with each entity treated as a client. Same workflow. Same controls. Single billing relationship.
What we believe
The choices that shape every line of the product.
Books reconcile, or the system does not post.
Every journal entry must satisfy SUM(debits) = SUM(credits) within ±0.0001. The check runs as a database CONSTRAINT TRIGGER, not in application code. Even an AI bug or a misbehaving service-role write is rejected at COMMIT time.
Posted is immutable.
Once a JE is posted, accounts and amounts never change. Corrections happen via reversal entries linked to the original, never silent edits. UI badges, audit log, and trial balance all show the pair.
Every field is traceable.
A trial-balance row drills into its journal entries; each JE drills into its source document; each document drills into the OCR/LLM job that produced it (model, prompt version, confidence). If the chain breaks, the field is wrong.
The AI is bound by the same rules as a person.
We do not give the model a service-role key. It reads what a junior bookkeeper would read inside the workspace, and nothing more. Customer Data is never used to train models — that prohibition is in our agreements with every AI subprocessor.
Compliance is encoded, not advertised.
SOCPA-aligned IFRS COA, ZATCA Phase 2 invoice extraction, GOSI salary templates, PDPL-aware regional hosting — these live in the schema and the prompts, not in marketing claims. We describe how we are designed and aligned; certification statements are added only when independent audits land.
Default-deny, every time.
If a feature would require leaking one client's data into another's view to make a workflow simpler, we do not ship it. We build the SECURITY DEFINER function that crosses the boundary safely instead.
The market we are building for
GCC accounting has a specific shape — bilingual, regulator-driven, paper-heavy, deeply human. We optimise for that shape, not for a generic accounting product translated.
Real workflows at firms across Saudi Arabia and the United Arab Emirates begin in Arabic and English on the same page, route through three or more approvers, end at a ledger that an auditor reviews in person against original PDFs, and must hold up under ZATCA, FTA, SOCPA, GOSI, and PDPL scrutiny. Tools written for North America or Europe and dropped into the GCC strip out half the workflow and replace it with friction.
We start from the workflow that exists. Voucher batches with cover sheets and supporting documents. Approval chains with named roles. Bilingual narratives. Period close with adjustment entries. Reversal-paired corrections. Audit-log-as-evidence. Then we layer agentic AI on top — extraction, classification, narrative generation, AI Inbox clarifications, document cross-check, dashboard digests — to compress the busy work that has historically eaten 70% of a junior accountant's week.
The win condition is not "replace the accountant". The win condition is "let the accountant spend 70% of their time on judgement and review, not data entry and re-typing." That is the wedge.
Who built it
A team of practising and qualified Chartered Accountants who got tired of watching juniors retype invoices.
Nexus Ledger is built by a team of Chartered Accountants — people who have actually closed bilingual books in the GCC, sat through SOCPA-aligned audits, defended trial balances against external auditors, and watched juniors burn weeks every month re-typing data that already existed in PDFs. We are building the tool we wished we had on the inside.
The product is operated under a parent holding company; the official corporate name and registered jurisdiction will be announced on incorporation. Until then, the Service is delivered under the Nexus Ledger brand and the contracting entity for pilots is the sole-proprietor operator named in the Terms of Service.
What we do not do
Boundaries we hold deliberately, because crossing them dilutes the wedge.
We do not auto-post.
During Phase 1 and during the 30-day cold-start window for every new workspace, every journal entry requires three human approvals before it posts. AI proposes; people decide.
We do not file your returns.
The Service prepares VAT registers, GOSI splits, ZATCA-aware extracts, audit packs, and trial balances. Filing those returns with ZATCA, the FTA, GOSI, or any other authority is the customer's responsibility.
We do not bundle ERP into accounting.
HR, Procurement, and CRM are real workflows but they are NOT part of the core accounting tier. They are opt-in add-ons, separately scoped and separately priced, with their own database schemas and RBAC. Customers who only want bookkeeping pay only for bookkeeping.
We do not use Customer Data to train models.
Not for our own models, not for our subprocessors' models. This is a contractual prohibition on every AI provider in our subprocessor list, and it is a precondition to processing.
Want to see it on real bilingual workflow?
The interactive demo is live and uses a fully anonymised reference dataset.