Tokenized Fund Operations on the Live Register

Kerovan keeps on-chain tokens and the legal register in step, and proves every fix.

When the token and the register stop agreeing

A tokenized fund keeps its records in two places. The tokens live on chain. The legal register, the transfer agent books and the bank ledgers live off chain. Every subscription, redemption, transfer and corporate action has to land in both places. When one side moves and the other doesn't, you get a break. Most teams find those breaks the slow way, in an end-of-day spreadsheet or a month-end reconciliation, or when an investor or auditor asks why the numbers don't match. By then the wrong holding may already have been traded, paid out or reported.

‍

The challenge: catch breaks live, and prove every fix

Fnd operations on a tokenized fund can't wait for a batch job. Token transfers settle in seconds, while the register and ledgers update on their own schedules, so the two drift apart all day. Fixing a break is a regulated act. It changes who legally owns what, so every decision has to hold up for DORA and MiFID evidence needs and for the fund's own auditor. And the data can't leave your perimeter to be checked by an outside service. So the job isn't only to spot a break in real time. It's to hold it, get it fixed by a person and prove afterward what happened and why.

‍

Why tokenized fund breaks stay hidden

Two ways to keep on-chain tokens and the legal register in step.

VS

End-of-day reconciliation

Breaks are found hours or weeks after they happen.

No proof of how a break was fixed
❌ Spreadsheets and batch jobs between chain and register
❌ Wrong holdings traded or paid out before anyone sees it
❌ Fixes made by hand with no audit trail

Checked on the live stream

Kerovan compares both books inside your network.

Every fix on the record
✅ Breaks caught as they happen and held for a person
✅ Fund and investor data never leave your perimeter
✅ Every decision logged for your auditor

‍

The Kerovan approach: agents on the live fund stream

Kerovan runs on Lascaris, the operating system for your agents. Its agents watch the token transfers on chain and the updates to the register, the transfer agent books and the bank ledgers, all as they happen. They match each movement against the other books. When a holding, a NAV or a transfer doesn't line up, they flag the break and hold it until a person approves the fix. Everything runs inside your network, on your own Kafka, so fund and investor data never leaves your perimeter. Every break, every decision and every fix goes into an immutable audit log, with who approved it and why. That record is what matters. Your operations, risk and audit teams can show exactly what happened to each holding, which is the evidence DORA and MiFID ask for.

‍

Use case: keeping a tokenized fund and its register in step

Here's the scenario. A fund manager runs a tokenized share class next to its traditional one. Subscriptions, redemptions and transfers move on chain all day. The legal register and the bank ledgers update on their own schedules. Most of the time both books agree. The one break that matters, such as a transfer to a wallet that isn't on the register or a NAV that doesn't match the price source, sits quietly until someone runs the end-of-day reconciliation.

‍

The Kerovan Reconciliation Loop

Every break caught and held for a person. Fund data never leaves.

On chain
Token ledger Live transfers
Data stays
Legal register TA and bank books
Break held for approval
Live fund stream

Kerovan agents

Every fix on record. Data never leaves.

‍

The agent solution. With Kerovan on Lascaris, agents run on the live fund stream next to your register and ledger systems. They match each token transfer, subscription and redemption against the register and the bank books. They link related movements, so a single event doesn't turn into five separate breaks. Anything that doesn't line up is held until a person approves the fix. Every break and every fix is recorded.

The result. Breaks are caught when they happen, not at month end, and every fix carries a recorded reason and approver. Fund and investor data never leave your network, and the whole process is ready for audit instead of being a black box.

‍

Every fix on the record

The hardest objection to automating reconciliation is the obvious one. What if the system changes a holding it shouldn't? Kerovan never writes back by itself. It reads both books, flags the break and leaves the decision to a person. Each step is stored as an immutable event, so operations, risk, the fund's auditor and the supervisor all work from the same evidence. For DORA and MiFID evidence needs, that trail is what lets you go live.

‍

Runs where your fund data lives

Kerovan connects to your existing Apache Kafka, Confluent or Redpanda, and to your register, transfer agent and ledger systems. If you don't run Kafka yet, Lascaris brings its own. It runs on-premise or air-gapped, and NAV and price data are signed at the source before they reach your oracle. Investor data stays inside your perimeter by design, not because a policy says so.

‍

Regulatory context

Tokenized funds now sit inside regulation in Europe. The DLT Pilot Regime lets market infrastructures run on distributed ledgers under supervision. DORA sets ICT risk and incident reporting rules for financial firms, and MiFID II sets record-keeping duties for investment firms. Any automation that touches the register also has to hold up to the fund's own auditor. Kerovan is built to meet that bar, not to work around it.

‍

Why this matters

Everyone knows tokens and registers drift apart. What's been missing is a way to catch each break as it happens and prove every fix. Kerovan provides that on Lascaris, without moving fund data or replacing your register or transfer agent systems.

Related: Kerovan for tokenized fund operations, Lascaris for the agent platform and audit trail, and the DORA and NIS2 checklist.

‍

External references:

About Scalytics

Scalytics builds Lascaris, the platform that runs governed AI agents on the Kafka you already have. We help banks, insurers and fund teams put AI into production without moving customer data and without losing the record of what each agent decided.

Our founding team created Apache Wayang, the open source framework that runs computation where the data lives instead of copying it around. The same idea is at the heart of Lascaris.

Our mission is to keep data in place, bring the compute to the data and give every regulated team AI they can defend in front of an auditor.

The experts in mission-critical data and AI.

Bring us your hardest problem. We'll scope it with you.