DIFC DIFC Innovation Licence · Dubai International Financial Centre

An AI-native platform for
financial operations, control and governance.

Reconciliation, an embedded ledger, compliance and audit — on one governed, multi-tenant platform. Built for finance shared services, internal audit functions and professional services firms across India and the GCC.

RLS PostgreSQL row-level tenant isolation
Iceberg open-format governed lakehouse
4 currencies — INR / USD / GBP / AED
14-day free trial, self-serve
The problem

Month-end runs on spreadsheets, email and trust.

In most mid-market finance functions, the controls that matter most are the ones held together by hand.

Reconciliation is manual and late

Bank, intercompany, AP/AR and control-account reconciliations are worked in spreadsheets, keyed line by line, and finished days after the period they describe. Breaks are found late, and the same breaks recur every month.

Controls leave no evidence trail

Approvals happen in inboxes and chat. When an internal auditor asks who reviewed a journal, or why a control was overridden, the answer has to be reconstructed from memory and forwarded threads.

Audit evidence is a scramble

Every review cycle re-opens the same hunt for support — the PBC list, the sample selections, the tie-outs. Evidence lives across drives and mailboxes rather than beside the transactions it supports.

Finance data is fragmented

The ERP, the bank portals, the sub-ledgers and a dozen spreadsheets each hold part of the truth. No single governed record ties a transaction from source, through reconciliation, to the ledger and the audit file.

Ainlytx addresses this as one connected pipeline — so a transaction is reconciled, posted, controlled and evidenced in the same governed system, not four disconnected ones.

The financial-control pipeline

One governed path from transaction to forward view.

Four modules move a transaction through reconciliation, the ledger, controls and audit — and a fifth turns that reconciled history into a forward-looking view. All of it runs on one governed data foundation.

01

Command Center Live

Transactions come in and are reconciled and transformed — bank, intercompany, AP/AR and control accounts matched, breaks surfaced, clean data prepared.

transactions in → reconciled
02

Mosaic Live

Reconciled activity is posted to the embedded accounting ledger — multi-entity, multi-currency, with an append-only, hash-chained journal.

reconciled → posted to ledger
03

GovernIQ Live

Controls run against the posted ledger — exceptions triaged, approvals captured, and audit evidence generated and held beside the transactions it supports.

posted → controls & evidence
04

Foresight In development — Q4 2026

Foresight will read the client’s own reconciled ledger data and compute forward-looking views — cash position, burn, runway and float — under assumptions the user sets.

controlled history → forward view
FOUNDATION

Vault Live

Every stage reads and writes the same governed data foundation — a per-tenant Apache Iceberg lakehouse. Reconciled outputs, ledger data and evidence are stored in open tables, isolated per tenant, and queryable by BI tools with a scoped token.

Foresight is in development, launching Q4 2026. It is the apex of the pipeline — the reason the earlier stages exist — not a standalone product. Its outputs will be informational and computed from user-supplied assumptions applied to the client’s own data. Foresight does not provide financial advice.
Why AI-native, specifically

The AI does the parts a reviewer can’t do by hand.

Concrete mechanisms on financial data — not a chat box bolted to the side of a ledger.

Anomaly detection in reconciliation

Statistical outlier tests, Benford’s-law deviation, duplicate and round-number detection, and weekend/backdated flags run across matched and unmatched items — so unusual entries surface without a human reading every line.

Exception triage

Unmatched and flagged items are grouped and ranked by likely cause and materiality, so a reviewer works the exceptions that matter first instead of an undifferentiated list of breaks.

Control & journal-entry testing

Journal-entry tests — round-trip detection, debit-equals-credit reversals, backdated and holiday postings, ghost-employee and duplicate-payment checks — run continuously against the ledger, the same tests an audit team would run at year-end.

Every one of these runs on the client’s own reconciled data, and every result is explainable to the reviewer who has to sign it off.

Platform & architecture

Built to pass a technical review.

The specifics a reviewer asks for, stated plainly.

Multi-tenant isolation

Every tenant-scoped table is protected by PostgreSQL row-level security, with tenant context bound to a short-lived JWT and enforced at the database. Application-layer filters sit on top as defence in depth. Isolation does not depend on application code being bug-free.

Open-format lakehouse

Governed outputs land in per-tenant Apache Iceberg tables. Access is granted per user, per table. BI and analytics tools connect over standard interfaces with a scoped bearer token — no proprietary driver, no lock-in.

SSO & identity

SAML and OIDC single sign-on with Azure AD (Microsoft Entra) and Okta, and any standards-based identity provider. Short-lived tokens, optional email one-time-passcode second factor, no shared service accounts.

Integrations

Connects with NetSuite, Tally, Zoho, SAP, Microsoft Dynamics, Google Workspace and Microsoft Graph, plus a REST API and webhooks. Bring source data in and push governed outputs to the systems your team already uses.

Security, data protection & residency

Where the data sits, and who can reach it.

A precise statement of the current model, with anything not yet in place marked as roadmap.

Tenant isolation

PostgreSQL row-level security on every tenant-scoped table, tenant context bound to the JWT, and per-table grants for lakehouse data. One tenant cannot read another tenant’s records.

Access control

SAML/OIDC SSO, short-lived tokens, optional email OTP second factor, and lakehouse access granted per user per table with one-click rotate and revoke.

Audit logging

Grants, logins, file actions, approvals and data-API calls are logged with user identity and timestamp, and retained for the configured retention period.

Encryption

Data is encrypted at rest with Google-managed keys and in transit with TLS. File uploads go directly to object storage over short-lived signed URLs.

Data residency

Model
Data residency follows the tenant’s location, held in the Google Cloud region in — or nearest to — the tenant’s country.
Region in use today
Tenant data is currently held in Google Cloud’s Mumbai region (asia-south1), in India.
India
In-country: data is held in the Mumbai region (asia-south1).
UAE & GCC
Google Cloud has no data-centre region inside the UAE, so we do not claim UAE in-country residency. For UAE and GCC tenants, data is held in the nearest region we operate — Mumbai (asia-south1), in India.
Other regions
On the roadmap — tenant-selectable regions for the EU and the US, so residency can follow the tenant into those jurisdictions. Not yet available.
Ainlytx holds a DIFC Innovation Licence. It is not regulated, licensed or authorised by the DFSA, does not hold any financial services permission, and is not a bank, payment provider or money services business. Ainlytx provides software; it does not hold or handle client funds.
Founder & domain credibility

Built by someone who has done the reconciliation, the ledger and the audit.

Founder & Chief Executive

Ramanpreet Kaur Bhatia

Chartered Accountant (India) · CPA Australia

  • Qualified Chartered Accountant (India).
  • Member, CPA Australia.
  • Hands-on background in reconciliation, financial reporting and audit — the workflows the platform automates.

Reconciliation, ledger and audit tooling is unforgiving to design if you have not done the work. The controls have to match how a real close runs, the evidence has to be what a real reviewer accepts, and the exceptions have to be ranked the way an experienced preparer would rank them.

Ainlytx is built by a qualified Chartered Accountant (India) and member of CPA Australia — someone who has sat on both sides of the review, preparing the numbers and testing them. That founder-market fit is the reason the pipeline is shaped around controls and evidence rather than dashboards, and it is the company’s strongest non-financial credential.

More about the company and the team →

Platform breadth

The same identity, access and data layer, extended to adjacent workflows.

Beyond the control pipeline, Ainlytx runs the operational workflows around a finance or professional-services team — on the same tenant isolation, access model and governed data plane.

Adjacent workflows for gym operations, India–GCC trade and CRM live on the platform extensions page.

DIFC & presence

Registered at the DIFC, building across three hubs.

DIFC hub — Dubai

Registered in the Dubai International Financial Centre under a DIFC Innovation Licence. This is an innovation-scheme registration — not a DFSA financial-services authorisation.

GovOps centre — Delhi

Engineering and technical operations run from Delhi, India — where the platform is built and where product and data engineering sit.

UK hub — planned 2026

A UK hub, Ainlytx Tech Solutions UK Ltd, is planned for 2026 to support customers in the UK and Europe.

Start on your own data.

Self-serve signup and a 14-day free trial, with published pricing in INR, USD, GBP and AED. Bring a real reconciliation or a month-end close and run it on the platform.