● Project Guardian is MAS's collaborative sandbox for asset tokenisation, testing tokenised funds, bonds, stablecoins, and bank liabilities with financial institutions, central banks, and regulators (mas.gov.sg).● No vendor or institution holds "MAS-approved" status in a formal certification sense. Participation runs through the Industry Group workstreams for asset management, fixed income, and foreign exchange, led by Schroders, ICMA, and Ant Group/ISDA.● Your infrastructure choice splits three ways. R3 Corda and Consensys are platforms you license and integrate with in-house engineers. Restart Fintech is the delivery partner that builds and ships a Guardian-aligned pilot for institutions without a blockchain team.

What Project Guardian Is and Who Can Participate

Project Guardian is a cross-border sandbox program, not a licensing regime. MAS runs it as a collaborative initiative between policymakers and the financial industry to test asset tokenization across tokenised funds, bonds, stablecoins, and bank liabilities. No institution earns a Guardian license, and no vendor receives a Guardian certification. Firms join by running industry trials that demonstrate commercial viability and regulatory alignment.

Two bodies govern the program. The Policymaker Group handles cross-border regulatory collaboration and includes MAS, Banque de France, Deutsche Bundesbank, the UK's Financial Conduct Authority, Japan's Financial Services Agency, the IMF, and the World Bank. The Industry Group brings together asset managers, custodians, market operators, and commercial banks, with members like BNY Mellon, Citi, DBS, Franklin Templeton, HSBC, J.P. Morgan, SGX Group, Standard Chartered, and UBS.

Actual work happens inside three named workstreams, each with a designated lead. Schroders leads Asset and Wealth Management, the International Capital Market Association leads Fixed Income, and Ant Group, with the International Swaps and Derivatives Association, leads Foreign Exchange. Tokenised bank liabilities for transaction banking sit under the FX workstream rather than a separate trade finance track, a distinction generic vendor lists often get wrong.

What real participation looks like is visible in the live use cases MAS cites as evidence of operational resilience. Ant Group, Apollo, DBS, Franklin Templeton, Hamilton Lane, OCBC, and UBS Group have all run implementations. These are the peer institutions a serious participant sits alongside, and each entered through a workstream aligned to its business rather than through a procurement badge.

The engagement path corrects a common misconception. MAS invites interested parties to register their interest through a link on the program page, and participation runs through the Industry Group workstreams and their named leads. There is no application scoring process, no accreditation prerequisite, and no third-party vendor-approval mechanism described in MAS's own material. An institution engages by registering directly and aligning with an existing member institution or workstream, not by hiring a company that holds a certification.

That absence of a certification scheme matters for anyone searching for "MAS-approved" providers. The right question is which workstream fits your assets and which participants you can align with, not which vendor carries an approval that does not exist.

Compliance and Technical Requirements to Work Within Guardian

MAS has not published a technical onboarding specification for Project Guardian, so no institution can point to an API reference, node requirement, or KYC/AML integration spec and treat it as the checklist. What exists instead is a set of published frameworks that function as the compliance reference material, and any institution planning a pilot needs to read them as design constraints rather than as a certification exam.

The Guardian Fixed Income Framework sets the clearest expectations. Its addendum covers delivery-versus-payment settlement for DLT-based debt securities and documents lessons from custody arrangements, and it folds in ICMA's Bond Data Taxonomy, CMTA's Token Standards, and GFMA's Design Principles for Tokenised Securities. An institution issuing a tokenised bond under Guardian is expected to show that its settlement achieves finality and that its custody model holds up under the same scrutiny a traditional custodian would face.

The Guardian Funds Framework extends the same logic to funds. It recommends how to tokenise the fund lifecycle, maintain on-chain share registers, and structure tokens under a proposed Composable Token Taxonomy. Read together with the fixed income framework, the practical requirement is consistency. Your custody, settlement, and token structure decisions have to map to a taxonomy other participants recognise, not to a bespoke design only your team understands.

Interoperability carries as much weight as custody or settlement. The Interlinking Networks and Enabling Open & Interoperable Networks frameworks, the second developed with the BIS Committee on Payments and Market Infrastructure, propose a common model for linking digital asset networks across different ledger technologies. An institution that builds a closed system compatible only with itself has satisfied none of Guardian's actual intent, because the program's stated objective is seamless issuance, trading, and settlement across platforms and jurisdictions.

Security and operational resilience run underneath all of it. MAS names them explicitly alongside custody and settlement finality as areas its risk and regulatory frameworks must address. The gap institutions must plan around is that MAS translates these into published principles, not a stamped-and-signed technical spec, so demonstrating compliance means interpreting the frameworks and defending your architecture rather than passing a defined test.

Methodology: How These Three Were Evaluated

Four dimensions decide whether an option fits a Guardian-aligned pilot, and they are the same four in the table above. The first is engagement model, meaning whether you license a platform and integrate it yourself or contract a partner to deliver the build. The second is technical architecture fit, since a permissioned ledger, an Ethereum-based stack, and a custom build each carry different tradeoffs for a given asset class. The third is participant profile match, because a large bank with a standing DLT team makes a different choice than a ministry with none. The fourth is regulatory and compliance support, which measures how much GFIF alignment, custody, and settlement-finality work each option carries for you.

Feature checklists miss all four. A Guardian pilot lives or dies on how the tokenized asset settles, who interprets the frameworks, and who writes the integration code, not on which vendor lists more capabilities.

Engagement model: platform license vs. managed delivery

R3 Corda and Consensys sell you infrastructure, not outcomes. Both operate as software platforms your own engineers license and integrate. R3 provides the Corda distributed ledger and its enterprise tooling, and Consensys ships Ethereum-based products for building tokenization and settlement applications. In either case, you sign a licensing or subscription arrangement, then your in-house blockchain team designs the network topology, writes the smart contract logic, connects to custody and settlement systems, and maintains the deployment. The platform vendor supports the software. Your team owns the build.

That model works when you already employ blockchain engineers who understand permissioned ledgers or the Ethereum stack. Most central banks, finance ministries, and mid-market financial institutions researching Guardian do not. They have compliance depth and domain expertise, but no standing team fluent in DLT node operations, DvP settlement logic, or token taxonomy design. Handing those institutions a Corda license or a Consensys toolkit leaves the hardest part unstaffed.

Restart Fintech occupies the delivery layer that sits between an institution and a platform. Rather than licensing you software to integrate yourself, Restart Fintech contracts fractional CTO-led delivery and builds the pilot. The engagement covers architecture decisions, custom implementation, and hands-on regulatory and technical work through to a working Guardian-aligned pilot. When a project calls for Corda or a Consensys component, Restart Fintech integrates it on your behalf rather than shipping it as a manual for your absent engineering team.

The distinction decides who does the work. With Corda or Consensys, you buy the tools and supply the builders. With Restart Fintech, you contract the builders and get a delivered pilot. An institution with an existing DLT team can go direct to a platform. An institution without one needs the delivery layer to turn a platform into a running system.

Technical architecture: permissioned ledger vs. Ethereum-based vs. custom build

R3 Corda and Consensys start from opposite architectural assumptions, and that divergence shapes which tokenization problems each one solves cleanly. Corda grew out of regulated financial market infrastructure, where every participant is known, and each transaction stays visible only to the parties involved. Its permissioned ledger records agreements between named counterparties rather than broadcasting state to a shared global chain, which maps closely to how custody, settlement, and bilateral clearing already work in institutional finance.

Consensys builds on Ethereum, and its enterprise tooling brings the smart contract model and the surrounding developer ecosystem to institutional projects. That gives you programmable composability and a large body of existing standards for issuing and moving tokens, at the cost of running your controls inside an environment designed for open, general-purpose computation. For tokenization projects that lean on composable token logic or need to interoperate with public Ethereum liquidity, the Ethereum-based path removes work you would otherwise rebuild from scratch.

Here honesty matters more than a clean comparison. No independent, non-vendor source in the published Guardian material confirms a direct R3 Corda or Consensys pilot, participant slot, or named case study. MAS's own Guardian page lists neither company in its Policymaker Group, Industry Group, or workstream leads, and the named live use-case participants are institutions like Ant Group, Apollo, DBS, Franklin Templeton, Hamilton Lane, OCBC, and UBS (mas.gov.sg). Treat the strengths above as properties of each platform category, not as evidence of a confirmed Guardian deployment.

A custom implementation approach sits across both, because Guardian's published frameworks care about outcomes rather than a specific ledger. The Interlinking Networks paper and the CPMI-developed interoperability work assume different institutions will run different ledger technologies and still need to settle across them (mas.gov.sg). Restart Fintech works in that gap. Rather than pushing one base layer, its fractional CTO-led delivery selects and integrates whichever architecture fits your asset class and counterparties, then builds the settlement, custody, and interoperability logic on top that a raw platform license leaves to you.

Participant profile: who each option actually fits

R3 Corda fits large banks and financial market infrastructure operators that already run internal DLT teams. The named Guardian participants set the realistic peer group here. DBS, OCBC, and UBS Group have run live tokenization pilots inside the program, and institutions of that scale typically employ engineers who can integrate a permissioned ledger, write settlement logic, and operate nodes without outside delivery support. If your firm looks like DBS, Corda's regulated-market heritage is a natural match.

Consensys-style Ethereum tooling fits projects that are already Ethereum-native or committed to public-chain interoperability. Ant Group, Apollo, Franklin Templeton, and Hamilton Lane have built Guardian use cases spanning tokenized funds and FX, and asset managers pursuing composable, cross-network token standards tend to favor an Ethereum base. That path still assumes in-house blockchain engineers who can deploy and secure the smart contracts themselves.

Central banks, finance ministries, and mid-market financial institutions sit in a different position. They carry the regulatory mandate to run a Guardian-aligned pilot, but they rarely staff a standing blockchain engineering team, and hiring one to deliver a single pilot is slow and expensive. That is the gap Restart Fintech fills as a fractional CTO-led delivery partner. Rather than licensing a platform and then searching for engineers to integrate it, you contract the engineering and regulatory delivery directly and ship the pilot with the team you already have.

The peer set makes the split concrete. A DBS or a UBS chooses infrastructure and builds it internally. A ministry evaluating tokenized bond issuance against the Guardian Fixed Income Framework, without an internal DLT bench, needs someone to interpret the framework and deliver working infrastructure. Match yourself to the peer group you actually resemble, not the one you aspire to fund a team for.

Regulatory and compliance support: what each option actually delivers

Guardian compliance is not a checkbox a vendor ticks for you. It requires interpreting the Guardian Fixed Income Framework, its DvP settlement guide for DLT-based debt securities, and the custody lessons published alongside it, then translating those into how your specific pilot settles, custodies, and reports. R3 Corda and Consensys hand you infrastructure that can be configured to meet those expectations, but neither reads the GFIF for you or decides how your custody arrangement satisfies settlement finality requirements.

Corda gives you a permissioned ledger with settlement-finality properties that map cleanly onto regulated market infrastructure, which shortens the technical argument for DvP compliance. The interpretive work stays with your team. You still map ICMA's Bond Data Taxonomy and CMTA's Token Standards onto your instruments and defend those choices to a workstream lead.

Consensys provides Ethereum-based tooling with strong tokenization primitives, and its compliance depends heavily on how you configure permissioning, custody integration, and reporting on top of it. The platform supplies the components. Your engineers assemble them into something a MAS Industry Group member will accept, and your legal and operations teams own the GFIF alignment.

Restart Fintech carries that interpretive load as delivery. Its fractional CTO model treats GFIF alignment, DvP settlement design, and custody arrangement as work to be delivered, not documentation to be handed off. For a central bank or ministry without an in-house blockchain team, that difference decides whether a pilot ships. A platform gives you a ledger. An implementation partner reads the frameworks, chooses the architecture that satisfies them, builds it, and stands behind the regulatory reasoning when the workstream lead asks how your pilot achieves settlement finality.

MAS Regulatory Requirements for Project Guardian Participation

No blockchain vendor holds an MAS approval badge for Project Guardian, and any provider claiming one is misrepresenting how the program works. MAS runs Guardian as a collaborative sandbox structured through Industry Group workstreams, not a licensing regime with a certified vendor register. The (mas.gov.sg) describes no application scoring, no accreditation mechanism, and no published list of approved technology providers. Institutions researching "which vendor is MAS-approved" are asking a question the program does not answer.

The better question is what your institution must demonstrate to participate credibly. MAS and industry groups have published the frameworks that function as the working compliance reference. The Guardian Fixed Income Framework now carries an addendum covering delivery-versus-payment settlement for DLT-based debt securities and lessons from custody arrangements. The Guardian Funds Framework sets out recommendations for tokenising the fund lifecycle, on-chain share registers, and a proposed composable token taxonomy. Meeting these means proving your custody model, your settlement finality guarantees, and your interoperability approach hold up under regulatory scrutiny.

Interoperability is where MAS has been most explicit. The Interlinking Networks framework proposes a common model for connecting digital asset networks across different ledger technologies, and Enabling Open & Interoperable Networks, developed with the Bank for International Settlements' CPMI, sets design principles for networks that can settle across jurisdictions. An institution that builds a closed system unable to link to other Guardian participants has missed the point of the program.

For buy-side institutions, Bridging the Adoption Gap matters most. IMAS and the UK's Investment Association developed it with MAS and FCA support, and it translates commercial, operational, legal, and regulatory expectations into actionable steps. It exists because asset managers kept finding that a technically sound token still failed buy-side requirements on structure and reporting.

One gap you must plan around directly. MAS publishes no technical onboarding specification, meaning no defined API standard, node requirements, or KYC/AML integration specs for firms wanting to build compatible infrastructure. You cannot download a checklist and integrate against it. You demonstrate alignment through the frameworks above and through the workstream you join, which puts the interpretation burden on your technical team.

Choosing a Platform vs. an Implementation Partner

Start with an honest audit of your engineering bench. If you already run a blockchain team that has shipped production DLT applications, licensing Corda or integrating Consensys tooling is the right move, and you do not need a delivery partner to hold your hand through it. Large banks and financial market infrastructures in the Guardian Industry Group, like DBS, UBS, and Apollo, fit this profile. They have the internal capacity to map GFIF requirements to code and to run a pilot themselves.

Contract Restart Fintech when you have a mandate to deliver a Guardian-aligned pilot but no in-house blockchain engineers to build it. Central banks, finance ministries, and mid-market institutions often sit here. They understand tokenization at a policy level, and they cannot justify a permanent blockchain hire for a single pilot, yet they still need someone to translate custody, settlement finality, and interoperability requirements into working infrastructure.

The distinction is not platform versus platform. Restart Fintech's fractional CTO model can build on Corda, on Ethereum-based tooling, or on a custom stack, then own the regulatory interpretation and technical delivery through to a running pilot. A platform license gives you infrastructure. It does not give you the person who decides how that infrastructure satisfies the GFIF DvP settlement guide.

Your Monday-morning test is simple. Can your current team, without new hires, stand up a tokenized asset pilot that maps to a published Guardian framework in the next two quarters? If yes, go direct to a platform and use your budget on licensing and integration. If no, engage an implementation partner first, decide the platform second, and let the delivery team recommend the architecture that fits your asset class rather than committing to a stack before you understand the compliance work it demands.

FAQs