Vision for Kakarot as a zkEVM: A Canary release for mainnet's EVM

[pedrobergamini]

During today's community call we discussed a few ideas around the implementation of account abstraction and the RPC client for Kakarot. I believe that Kakarot may be useful for not only working as a solid type 3 zkEVM that shows all the power of Cairo and StarkNet features, but also as a canary release of the mainnet EVM.

We have the freedom of writing opcodes and EIPs that have not been implemented in Ethereum yet, and through Kakarot's RPC devs can easily integrate with it using popular tools such as Hardhat and Foundry. By having a roadmap that elects potential EIPs to be implemented in mainnet, or simply EIPs that have a significant dev adoption and interest such as EIP-1153 for example, we can solve two relevant problems in the ecosystem with Kakarot:

• Having a solid, functional and friendly implementation of a type 3 zkEVM
• Allow Ethereum community developers to test novel protocols using new EIPs or opcodes in Kakarot before it's available on Mainnet

Would like to get the community's opinion and whether it goes too off topic of Kakarot's vision. I'm also available to discuss and help implement some interesting EIPs maybe in a separate branch so we can test the concept.