L2 tooling and UX push
Several layer‑2 projects reported rapid product advances this week—Taiko cut block proposal costs by 22x, Aztec announced a full privacy stack, and Morpho deployed AI agents—all part of a broader sprint on L2 usability and features (x.com). Meanwhile, sequencing and latency solutions like Puffer are trialing preconfirmation paths that aim for sub‑100ms guarantees on OP‑Stack rollups, shifting the emphasis from raw scaling to app‑level user experience (x.com).
Ethereum’s layer-two networks are spending April on product plumbing, not just throughput, with new releases aimed at cheaper transactions, private apps, and faster confirmations. (morpho.org) (launch.aztec.network) (coinmarketcal.com) A layer-two network is a system that processes activity off Ethereum and then settles the result back to Ethereum, like a side road that still ends at the main highway. Taiko said its Shasta mainnet upgrade went live on April 9 and cut block proposal costs from about 1,000,000 gas to 45,000 gas, a roughly 22 times reduction, while proof costs fell from 500,000 gas to 280,000 gas. (taiko.mirror.xyz) (coinmarketcal.com) Privacy on Ethereum has usually meant hiding a transfer, not hiding what an application does. Aztec’s Alpha launch page calls its release a “feature complete privacy stack,” and outside analysis dated April 8 said the Alpha Network launched on March 31 with private execution, private identity, and private data built into the contract layer. (launch.aztec.network) (alearesearch.io) Aztec said developers can write contracts in Noir, mark functions public or private, and let private functions run on a user’s device before proofs are sent to the network. Alea said only proofs and public state touch Ethereum, with batches settling roughly every 12 seconds. (alearesearch.io) (binance.com) Morpho took a different route on April 8 by shipping tools for software agents rather than human users. In its beta announcement, Morpho said its new command-line interface and machine-control server give artificial intelligence agents read, simulate, and write access to Morpho lending markets across Ethereum and Base, and that every write action is simulated before execution. (morpho.org) Morpho said the package works with wallets from Coinbase, Safe, Fireblocks, and local keystores, and named Claude, Cursor, Codex, Windsurf, and more than 30 other agents as compatible tools. The company also said more than 130,000 artificial intelligence agents have registered onchain identities since January, citing Dune data. (morpho.org) The speed push is moving to the part of a rollup that decides transaction order and tells users a payment is effectively on the way. Puffer says its UniFi active validation service is a preconfirmation layer for rollups, advertises sub-second transaction confirmations on its site, and separately describes a roadmap for 100 millisecond transactions on Ethereum-linked rollups. (puffer.fi) Taiko has been arguing for the same shift in emphasis for months. In a February 6 post, Taiko said it was building “based preconfirmations” to improve speed and user experience on its rollup, after positioning itself as a “based rollup” that removes the need for a centralized sequencer. (taiko.mirror.xyz) The common thread is that layer-two builders are now changing the parts users actually feel: fee overhead, privacy controls, wallet flows, and waiting time after a click. The releases this month did not change Ethereum’s base rules, but they did push more of the competition into software design above the settlement layer. (coinmarketcal.com) (alearesearch.io) (morpho.org) (puffer.fi)
