Cartesi is a layer-2 infrastructure that creates a bridge between Linux and blockchain technologies. It gives decentralized application massive processing power and industrial-grade programmability. Complex and intensive computations run off-chain, while retaining the security guarantees of the blockchain.

With Cartesi, developers can continue to use the mature software infrastructure they are already proficient with, supported by Linux. They are free from the hassle of limited idiosyncratic blockchain environments. Instead, they can build on decades of software development and combine existing components to build their blockchain applications. Mainstream developers encounter a welcoming, familiar environment on which to express their ideas.



The Cartesi founding team initially got into blockchain as a result of Augusto Teixeira (Chief Scientific Officer) being a colleague in research and having co-authored scientific publications with Serguei Popov (author of the Tangle specification behind IOTA). As Augusto is an expert in probability and graph theory, with strong game theoretical background, Serguei Popov encouraged him to contribute to the blockchain movement by first introducing him to Bitcoin‘s whitepaper. Before the scope of Cartesi was defined as a generalized Linux infrastructure for scalable DApps, the original idea was to implement a decentralized data science and machine learning marketplace on the blockchain. This application required a deterministic off-chain compute infrastructure. To build that, Augusto needed more than his own skills. For this reason, he decided to involve Diego Nehab (Chief Technology Officer).

Diego and Augusto decided to on-board Erick de Moura as a CEO due to his close friendship with Diego (used to be roommates in UCLA in 1998) as well as the fact that Erick is a serial entrepreneur with experience in being the lead engineer in several tech companies and hands-on in software architecture and development for 20 years, having previously worked on his own blockchain projects.

Colin first got involved in blockchain through the venture capital world, working closely with portfolio companies who were building exchanges for cryptocurrency trading, and existing companies implementing blockchain technology. Their connection to Cartesi was the result of a close friendship with Erick and a passion for building a piece of the web 3.0.

Augusto and Diego started their research and early coding around August 2017. The research, the maturity of concepts and coding progressed steadily since the beginning. Early in 2018, Cartesi set its clear vision and purpose as a decentralized Linux infrastructure for scalable blockchain applications.


Cartesi Core

Cartesi Core is an infrastructure comprised of on-chain and off-chain components. In a nutshell, each participant that wants to interact with a Cartesi DApp does so through the off-chain component, which is called the Cartesi Node.

By moving computation off-chain, Cartesi Core removes the processing load from the consensus layer while preserving the security guarantees of the underlying blockchain. That’s rendered possible by the combination of three principles.

A Fitting Virtual Machine Architecture. Cartesi includes a custom VM, called the Cartesi Machine, that deterministically emulates a RISC-V microprocessor and runs an embedded Linux distribution. That’s where the decentralized logic of the DApp is executed. The Cartesi Machine is self-contained and reproducible. These two properties are necessary for DApp participants to retain verifiability and achieve consensus.

A Cryptoeconomic Protocol

Cartesi uses economic incentives to encourage DApp participants to engage with the system honestly and in the most efficient way, by responding timely and achieving common agreement on the results of computations that happen off-chain.

On-chain Dispute Resolution

If DApp participants fail to achieve agreement off-chain, the blockchain becomes a supreme court meditating a dispute resolution protocol. Bad actors suffer heavy economic punishments, while honest participants are economically compensated by the inconvenience suffered. Even in the rare circumstances when arbitration is needed, the verification process is performed with an interactive protocol that involves a negligible amount of computation and cost for the blockchain.

Data Ledger

While the Cartesi Nodes greatly expand the computational limits DApps, other scalability problems remain such as the high cost of transactions and the issue of data availability. To mitigate these problems, while maintaining the system portable across blockchains, Cartesi will develop a decentralized side chain based on Proof of Stake.

Cartesi’s Data Ledger will be optimized to work with the core technology of the Cartesi Machines. It will be built for short term storage, garbage collection, sharding, offchain emulated computations and localized consensus.

Data Ledger will employ a simple and yet robust implementation. This is possible as the main chain will be used in various critical parts of the consensus algorithm, such as random number generation and block ordering.


Cartesi reserves a 250 million CTSI allocation to a mine. The mine reserve will be used to generate extra incentive for early adopters and miners, helping to bootstrap the Cartesi Network. These tokens will be distributed to Cartesi Nodes being selected by Cartesi Network’s PoS algorithm used by the Data Ledger.

The CTSI rewards that miners will obtain from the reserve will gradually reduce with time. That will happen similarly to the way that miners of Bitcoin or Ethereum obtain block rewards, with a geometric decay along the years. These reducing rewards will be offset by the fees collected from Cartesi users for the services provided by the network.

CTSI Utility Token

Cartesi is currently developing a Proof of Stake blockchain running on top of Ethereum and maintained by a network of Node Operators. The Cartesi Token will be essential for this system to work as it provides:

  • Staking — Since block generators will be selected in proportion to their stakes in the economy, there needs to be a way to query and lock everyone’s balance in the system.
  • Election of block producers — The miners will be selected to propose blocks in proportion to their token balance, so that the network is run by a pulverized community of stakeholders.
  • Slashing — Bad behavior is disincentivized by locking and potentially slashing the tokens of dishonest actors.
  • Transaction fees — Although we are designing our Side Chain to be orders of magnitude cheaper than the underlying main chain, users or DApp developers still need to pay for their transactions to be processed, transmitted and temporarily stored by others. These fees will be paid to Node Operators with the Cartesi Token.
  • Challenge computations — The Cartesi Token will also be important in order to challenge computation results posted on the blockchain. This will work as an exit mechanism that guarantees the correct execution of DApps and discourages bad behavior.

External Links