Marco Argentieri – Arkade: A Virtual Execution Layer for Bitcoin-Native Applications

I’m Marco Argentieri, CEO and co-founder of Arcabs, and I’ve spent the last two years building the ARC protocol. At BTC Prague, I shared why virtualizing the UTXO is the key breakthrough Bitcoin needs to unlock true horizontal scalability — and what we’re building with Arkade to make it real.

We’ve Been Here Before

Think back to the early 2000s. If you wanted to build a web application, you had to physically install a server in a data center. One company, one server. When Christmas came and traffic spiked, you had to vertically scale — get a bigger machine. That didn’t work. Then in 2006, Amazon launched EC2 and suddenly you had virtual machines that looked like servers but scaled horizontally. That unlocked a trillion-dollar internet industry. I believe we are at exactly that inflection point with Bitcoin right now.

The Problem: Bitcoin’s Mempool Is Empty — But Not For Good Reasons

Bitcoin is being used more than ever, but activity is happening off-chain — through ETFs, custodians, other blockchains. People left because the UTXO felt like that physical server: expensive, slow, inflexible. And when the mempool spikes, businesses can’t tell customers “sorry, come back next month.” That’s a fundamental problem you can only solve with true horizontal scalability.

Virtual UTXOs: The Core Breakthrough

What ARC does is take the UTXO — Bitcoin’s core primitive — and virtualize it. A virtual UTXO (vUTXO) gives you full self-custody. You can always unilaterally exit. You don’t trust a third party. But at the same time, it’s faster, cheaper, and more flexible. Instead of batching 100 outputs into one transaction like Coinbase or Binance does today, ARC nests potentially tens of thousands of virtual transaction outputs under a single on-chain output. More people, lower cost — that’s real horizontal scaling.

Arkade: Virtualizing the Mempool Too

With our implementation, Arkade, we went further. We virtualized the mempool itself. This gives us instant pre-confirmation: you and the server co-sign and advance a sequence of transactions with a timestamp right now. We add security using secure enclaves — the same technology banks use — to guarantee pre-confirmations can’t be double-spent. The operator’s key is generated inside the enclave, inaccessible to any human, and communication is end-to-end encrypted.

We also flip the fee model. Today the sender decides the fee and the recipient just hopes it confirms. In Arkade, the recipient decides — pay now for the next block, or wait until next week when fees are lower. That flexibility is what businesses actually need.

Use Cases That Matter

The first obvious use case is Lightning payments via swaps — executed as first-class contracts inside Arkade, non-custodially, without complex federation setups. Then there are loans: people want to borrow against their Bitcoin without selling it. If we ever hit banking scale — thousands of loans per second — the base layer can’t handle it. People will leave again. Arkade can handle it. Escrow marketplaces, betting contracts, AI agent micropayments — these are all coming, and Bitcoin’s UTXO model with parallel execution is perfectly suited for them.

We’re also adding extended op codes in the virtual mempool — so Arkade supports covenants, introspection, lending pools, trading pools, non-interactive swaps. All built on basics that already exist in Bitcoin today: PSBTs, Taproot — nothing more, nothing less.

It’s Already Real — Come Build

We’ve shipped three SDKs in JavaScript, Rust, and Go. We have an open-source MIT-licensed wallet on testnet right now. Someone at the conference told me they built a wallet in