Smart Wallet & Account Abstraction

Problem

EOA-only wallets force every user into the same signing model: one key, one signature, gas paid in the native asset. ERC-4337 decouples the account from the signer, enabling gas sponsorship, batched operations, and social recovery — but the UserOperation flow, bundler/EntryPoint/Paymaster separation, and competing account implementations (Safe, Kernel, Biconomy) are easier to misunderstand than to use. A working proof-of-concept was needed to ground product decisions at Graylab in firsthand behavior rather than spec reading.

Approach

• Build against real bundler infrastructure rather than mocked EntryPoints, since the bundler-mempool interaction and Paymaster validation rules are where most implementation surprises live.
• Compare multiple smart account implementations (Safe, Kernel, Biconomy) to identify patterns worth adopting versus protocol-specific quirks.
• Evaluate signature aggregation with multi-sig and session keys in mind, since both are primary drivers of real product demand for AA.
• Contrast with Klaytn’s native account-update model to understand tradeoffs between protocol-level account flexibility and contract-level abstraction.
• Wagmi + Next.js for the client so the demo matches the stack used by downstream product teams.

Implementation

ERC-4337 Standard

• Studied the UserOperation flow: bundler → EntryPoint → wallet contract
• Analyzed Paymaster patterns for gas sponsorship and ERC-20 gas payments
• Evaluated different smart account implementations (Safe, Kernel, Biconomy)
• Researched signature aggregation for multi-sig and session key use cases

Smart Wallet Proof-of-Concept

• Built smart wallet interface using Next.js and Wagmi
• Integrated with ERC-4337 bundler infrastructure
• Implemented custom EntryPoint and account contract interactions
• Tested batched transaction execution and gas estimation

Klaytn Account System

• Researched Klaytn’s native account update mechanism
• Implemented account key rotation and multi-sig configuration
• Compared Klaytn’s account model with ERC-4337 approach

Outcome

• Working proof-of-concept demonstrating gas sponsorship, batched transactions, and social recovery under ERC-4337.
• Comparative analysis of Safe, Kernel, and Biconomy smart account implementations.
• Side-by-side evaluation of ERC-4337 versus Klaytn’s native account-update model.
• Hands-on findings that informed wallet product development decisions at Graylab.

Technologies

• Smart Contracts: Solidity, ERC-4337 EntryPoint
• Frontend: Next.js, TypeScript, Wagmi
• Networks: Ethereum, Klaytn
• Infrastructure: Bundler, Paymaster