Financial infrastructure requires rethinking blockchain architecture

The crypto industry has an infrastructure problem that’s rarely discussed directly: we’ve been building financial systems on blockchains that weren’t designed for finance, which requires us to rethink blockchain architecture.

I noticed this the moment we started building Momentum. Most protocols launch as isolated products, a DEX, a lending market, a staking solution, treating each as a separate tool rather than part of an interconnected system. But this fragmentation reveals a deeper architectural mismatch. The blockchain layer underneath simply wasn’t built to handle what finance demands: parallel processing at scale, composable primitives, and infrastructure that other protocols can reliably build upon.

This isn’t theoretical. It manifests in transaction failures during peak demand, capital inefficiency in liquidity markets, and an ecosystem where each protocol operates in isolation rather than synergistically.

You might also like: The next phase of onchain finance needs regulatory infrastructure, not just issuers | Opinion

The real constraint: Blockchains weren’t designed for finance

When we were deciding where to build our DEX, the choice was obvious to me but seemed counterintuitive to many. Everyone asked: Why not Ethereum (ETH)? The answer reveals everything about how I think about infrastructure.

Consider the fundamental difference between how Ethereum and Sui (SUI) process transactions. Ethereum’s sequential execution model means every transaction must be processed in order, creating bottlenecks under load. This wasn’t a bug in Ethereum’s design; it was never the intended use case. Ethereum was built to be a general-purpose compute platform.

Finance demands something different. Most financial operations are independent. When Alice swaps tokens and Bob stakes assets, these transactions don’t depend on each other. Sequential processing creates artificial congestion. Parallel processing is not just an optimization; it’s structurally necessary.

Sui was built from the ground up with parallel execution and object-centric design using the Move programming language. This architectural choice isn’t just faster — it enables an entirely different category of financial products to exist at scale.

The proof came faster than we expected. In six months, our DEX scaled from zero to $500M in liquidity and $1.1B in daily trading volume, accumulating $22B in cumulative trading volume while onboarding 2.1 million users without meaningful congestion. Processing that kind of volume without transaction failures isn’t a marketing achievement; it’s evidence of fundamental architectural soundness. Try achieving those metrics on a sequentially-executing blockchain and you’d see exactly why the architecture matters.

Why infrastructure composability matters more than individual products

There’s a second, more subtle problem I’ve learned to recognize: financial products should be composable building blocks, not isolated silos.

A properly designed financial infrastructure layer should allow other protocols to build on shared primitives. If every protocol has to build its own treasury management, its own staking solution, its own liquidity infrastructure, the ecosystem fragments. Developers spend time solving identical problems rather than innovating on new ones. I’ve watched this happen repeatedly across chains.

This is where most protocols fail. They build one product well, then the ecosystem around them calcifies. Each new protocol essentially starts from scratch.

When we built our protocol, we deliberately chose not to just create a DEX. We built infrastructure primitives that other protocols would rationally choose to use rather than rebuild. MSafe, our treasury management solution, now secures hundreds of millions across the Move ecosystem. Not because we forced adoption, but because it solved a real problem better than the alternatives.

More protocols building on shared infrastructure means more integration points, more composability, and higher system value for everyone. This only works if the primitives are actually good. Concentrated liquidity market-making technology with aligned incentives creates capital efficiency that traditional AMMs can’t match. Liquid staking that produces a yield-bearing receipt token creates collateral that’s simultaneously productive. Multi-signature treasury management that works reliably reduces friction for protocol governance.

These aren’t nice-to-have conveniences. They’re the difference between an ecosystem that compounds value and one that fragments. This is precisely what allows Momentum to provide infrastructure that other protocols rationally choose to build on rather than rebuild themselves.

The institutional capital problem is infrastructure, not features

Crypto has always struggled with institutional adoption. The standard explanation focuses on regulatory uncertainty or UX limitations. The real bottleneck is often simpler: institutions can’t use decentralized infrastructure that lacks compliance capabilities.

This isn’t a reason to centralize. It’s a reason to build the right layer on top of decentralized infrastructure. If you can offer permissioned compliance as an optional module, let institutional users verify their identity and trade with full regulatory clarity, while keeping the base infrastructure permissionless, you solve the problem without compromise.

Institutions won’t deploy serious capital into systems that can’t provide regulatory auditing, KYC verification, or compliance documentation. These aren’t features, they’re structural prerequisites for institutional participation. That’s not gatekeeping. It’s acknowledging reality.

The actual argument

Here’s the claim I’m making, separate from any particular protocol: Blockchains built for general computation cannot efficiently serve as financial infrastructure. Finance requires architecture specifically designed for parallel processing, composable primitives, and institutional compliance. Protocols will migrate toward blockchains with these properties—not because they’re trendy, but because the economics of operating on better infrastructure are simply superior.

This isn’t an argument that “Sui is better than Ethereum.” Ethereum can and should continue evolving. Layer-2 solutions are legitimate approaches. This is an argument that financial systems need to be built on different architectural foundations than general-purpose compute platforms.

The corollary is less obvious: if a blockchain is purpose-built for finance and achieves meaningful adoption, it becomes the natural foundation for financial innovation. Not because of marketing, but because other protocols rationally choose to build there.

The question for the industry isn’t which chain “wins.” It’s whether we’re willing to acknowledge that one-size-fits-all blockchain architecture was never the right approach, and that specialized infrastructure produces better financial outcomes.

That realization changes everything about how protocols should be built and where they should be deployed. It’s changing how I think about Momentum, and it should change how you think about where to build next.

Read more: While tokenization makes headlines, infrastructure will decide who wins | Opinion