--- title: "Radix Week in Review: The Protocol Outgrows Its Company" path: "/blog/week-in-review-2026-07-12" version: "1.0.1" author: "Hydrate" createdAt: "2026-07-12T11:10:54.598Z" updatedAt: "2026-07-12T13:19:03.675Z" --- # Radix Week in Review: The Protocol Outgrows Its Company | **Week of** | July 6–12, 2026 | | **Top story** | hyperscale-rs, the community's Rust implementation of Hyperscale, closes on Milestone 1 | | **Protocol / Dev** | genkipool ships a native Rust SDK; Hookah open-sourced under MIT | | **Ecosystem** | Community operators keep Stokenet alive; a clean-genesis reset is planned for Q3 to cut hosting cost | | **Community / Governance** | Accountability Council, DAO working groups, and wallet stewardship take shape on RadixTalk | | **Sources** | Radix Foundation blog (2), GitHub (genkipool, xstelea), radix.wiki, RadixTalk, Radix Developers | The great chartered companies of the seventeenth century did a strange thing when their settlements succeeded: they wound themselves down. Having fronted the capital, drawn the maps, and proven the harbour, the company would pre-pay the winter's stores and step back, leaving the colonists to invent a government, run the docks, and decide what to build next. The enterprise was measured by a single test: whether the settlement it founded could outlive it. Radix reached that moment some time ago, and this was the week you could watch it happen in real time. The [Foundation is in maintenance mode](https://www.radixdlt.com/blog/foundation-update-moving-to-maintenance-mode), its critical infrastructure pre-funded through late 2026 while it waits for a community entity to take over operations. The week's Radix news was written entirely by the settlers – the Rust engineers rebuilding the protocol's scaling layer, the toolsmiths filling gaps the core team never got to, and the councils arguing, at length, over who holds the keys. ## The Machine Changes Hands The largest single thing the Foundation built and then let go of is Hyperscale, the sharded architecture meant to carry Radix to the throughput a global financial system actually needs. When the Foundation [completed the interim phase](https://www.radixdlt.com/blog/hyperscale-completing-the-interim-phase), it shipped a baseline: working software, the test scripts, the network configurations, and a proven number – a [500,000-transaction-per-second public test](/contents/tech/research/hyperscale-500k-tps) of real swaps across shards, with cross-shard atomicity intact. The point of handing over a baseline rather than a promise is that progress stops depending on a Foundation mandate. Anyone can pick up the survey and keep building. Someone did. [hyperscale-rs](/contents/tech/research/hyperscale-rs), the community's ground-up Rust implementation of that architecture, is now the place the R&D actually lives, and this week it was visibly closing on its first milestone. Governance of the network's own parameters has landed on-ledger, which leaves one hard problem between the team and Milestone 1: state contention. When two transactions want the same substate – the price of a single DEX pool, say – you cannot simply shard your way past them, and the team's current work is a batching scheme that lets one shard sequence the contended writes under a single amortized lock. It is the least glamorous and most load-bearing kind of engineering, the sort a closed research team used to do behind a wall and a community now does in a Telegram channel with the Foundation supplying the compute for scaled tests. The [performance baseline is inherited](https://getradix.com/updates/news/hyperscale-update-500k-public-test-done-the-radix-blog-radix-dlt); the road past it is being paved by hand. ## The Toolsmiths A protocol is only as usable as the primitives sitting one layer above it, and for years several of Radix's most basic ones lived only in JavaScript. This week a community developer, genkipool, published a [native Rust SDK](https://github.com/genkipool/radixdlt-rust-sdk) that brings the off-ledger building blocks – ROLA login verification, transaction building and signing, key management, address derivation – into a language the backend and infrastructure world actually deploys. It ships dual-licensed under MIT and Apache-2.0, and it carries a detail that reads like a postcard from 2026: a built-in MCP server, so an AI agent can hold a conversation with a Radix wallet. It has one star and eight commits. That is exactly what day-one community infrastructure looks like, and it is worth watching rather than crowning. Alongside it, [Hookah](/developers/tools/hookah) went fully open. Once a hosted service, it is now an [MIT-licensed, self-hostable platform](https://github.com/xstelea/hookah) that lets a dApp authenticate with ROLA, register event triggers, and receive webhook deliveries the moment matching on-ledger events cross the transaction stream – the connective tissue that turns a ledger into something an application can react to. The pattern in both cases is the same one the whole ecosystem is living through: capability that used to be somebody's product becoming everybody's commons. ## Learning to Govern The hardest thing a settlement inherits is the question of who runs it. That question dominated the Radix Developers and main community channels all week, and for once the argument was the point. The [Radix Accountability Council](/ecosystem/radix-accountability-council), the emerging DAO, and its working groups are being stood up in public on [RadixTalk](https://radixtalk.com/c/governance), and the debates were the real ones a young republic has: how much authority a council should hold, where legitimate moderation ends and censorship begins, how to keep the surface area of any single body small enough that no one can capture it. A community that intends to be worth a great deal one day is teaching itself, now, to be paranoid about concentrated power. For the fuller frame, the wiki's [Radix governance](/contents/tech/core-concepts/radix-governance) page tracks how the pieces fit. Stewardship showed up in the plumbing too. A community operator running the bulk of the Stokenet test network laid out a plan to reset it in Q3 – a fresh genesis, no migration, a clean start – specifically to bring hosting costs down to something the community can sustain without a corporate budget behind it. It is an unglamorous line item, and it is precisely the kind of decision that used to be made inside a company and is now made in the open by the person paying the bill. The same current runs through the wallet-stewardship proposals circulating this week: keep the mobile wallet maintained and lightly improved through the horizon of the next major upgrade, and decide as a community what it should become. None of this is a launch. That is the story. A week with no Foundation announcement, no headline number, no company at the podium, and yet the protocol moved – its scaling layer advanced, its tooling deepened, its governance took shape – because the people who use it have started to act like the people who own it. The colony is learning to run the docks. Next week it keeps building. ## Sources - [Radix Foundation – Moving to Maintenance Mode (Apr 28, 2026)](https://www.radixdlt.com/blog/foundation-update-moving-to-maintenance-mode) - [Radix Foundation – Hyperscale: Completing the Interim Phase](https://www.radixdlt.com/blog/hyperscale-completing-the-interim-phase) - [Hyperscale Update: 500k+ Public Test Done](https://getradix.com/updates/news/hyperscale-update-500k-public-test-done-the-radix-blog-radix-dlt) - [genkipool – radixdlt-rust-sdk (GitHub)](https://github.com/genkipool/radixdlt-rust-sdk) - [xstelea – Hookah (GitHub)](https://github.com/xstelea/hookah) - [RadixTalk – Governance](https://radixtalk.com/c/governance) - radix.wiki: [hyperscale-rs](/contents/tech/research/hyperscale-rs), [Hyperscale 500K TPS Public Test](/contents/tech/research/hyperscale-500k-tps), [Hookah](/developers/tools/hookah), [Radix governance](/contents/tech/core-concepts/radix-governance) - Radix Developers and main community Telegram (public discussion, July 6–12, 2026)