Nammu21 aims to move syndicated loans onto Google Cloud's ledger

Manual processing slows the loan market

Syndicated loans and the private credit market play an important role in corporate financing, but the processes involved are often manual. In lending transactions, facility agents, borrowers and syndicates of multiple lenders must coordinate bespoke terms, changing details and complex covenants.

According to Google Cloud, the fragmented, manual interpretation of legal texts creates operational burdens and data silos. Based on data from the Loan Market Association, settlement in the secondary market for par or near-par loans took an average of 42 days in the fourth quarter of 2024, with a median of 27 days. Lengthy processes tie up capital, increase counterparty risk and create a significant coordination burden.

Legal contracts become programmable credit instruments

Nammu21's platform converts the terms of loan agreements into Digital Credit Instruments, or digital credit instruments. The system links the terms, approvals, eligibility criteria and operating rules back to the original legal provisions, then expresses them as executable behaviours based on the current state of the loan.

The proof of concept uses Google Cloud Universal Ledger, a permissioned distributed ledger designed for institutional use. According to Google Cloud, GCUL offers a high-performance execution environment, near-instant transaction finality and predictable costs. Programmable contracts are written in GCUL's own language, gculpy. This is a strictly typed subset of Python designed to handle secure, readable and auditable financial logic.

Nammu21 performs two tasks. First, the platform automatically creates the digital credit instrument from the original agreement, then derives gculpy contracts from it. These determine which actions are permitted, required or prohibited on GCUL.

The process was tested on a $1.945 billion loan

The proof of concept modelled a $1.945 billion Term Loan B transaction that formed part of a multi-facility deal. The agent bank, borrower, assignor and assignee worked from the same permissioned ledger view.

The process begins when the agent bank selects the loan agreement transformed by Nammu21. The system provides the terms in a structured format, from which the facility's bespoke gculpy contract and digital twin are created on GCUL. The borrower and lenders are then onboarded individually. Each participant reviews the request through their own interface and gives explicit approval.

The agent bank then records an immutable hash of the loan's key terms on the ledger. Subsequent events, such as secondary-market assignments and repayments, are also handled through automated, multi-step processes. Transferring a loan position requires the consent of the new lender, the borrower and the agent bank, while execution takes place atomically.

Interest payments and loan closures can be settled directly on GCUL using tokenised commercial bank money. Role-based access protects trading strategies, positions and sensitive loan data, while the contracts check every action in advance against the legal terms.

Participants can work under the same rules

Google Cloud and Nammu21 say the approach could reduce settlement times, back-office workloads and counterparty risk. Contractual logic runs directly on the ledger, so participants do not need to reconstruct the state of a transaction from separate interpretations and subsequent reconciliations.

The proof of concept demonstrates how the same architecture could be applied to loan transactions with bespoke terms that currently require manual interpretation. According to the source, shorter settlement cycles could free up capital tied up for longer periods, enabling a more liquid and efficient secondary credit market.