Using zk-VMS with edge computing architectures in Ethereum to enable end-clients to compute on their devices and submit proofs instead of having execution layer nodes run all the computations is what I think the end state of Ethereum will end up becoming. By having an Edge Computing Interface and making the gas cost dependent on the length of the output variables, you could address the challenge of unbounded computation in WASM.
Do you think integrating this approach into existing smart contract systems could pose challenges or is there a work around? I want to envision developers adapting their smart contracts to work with this Edge Computing Interface but wonder if they need to rewrite their contracts, or could there be a seamless transition?