GitHub has rewritten the entire GitHub Copilot agent runtime from TypeScript and Node.js into more than 800,000 lines of production Rust code, the Microsoft-owned company disclosed Wednesday. The migration, which took roughly 14 and a half weeks, was carried out primarily by a single developer using GitHub's own Copilot coding agents to handle most of the programming work. The company says it switched to Rust in pursuit of better memory safety and performance, joining a growing list of major tech firms rewriting core software in the systems programming language.
The conversion spanned 128 pull requests that were merged and shipped incrementally while the rest of GitHub's team continued expanding the runtime's capabilities, according to Microsoft engineer Stephen Toub. Bun, the JavaScript runtime acquired by Anthropic last December, announced in July that it was rewriting more than half a million lines of Zig into Rust, with founder Jarred Sumner using a pre-release Claude model to complete much of the translation in 11 days. OpenAI quietly revealed last Friday that two engineers, working with Codex and GPT-5.5, rewrote Habitat—the storage service powering ChatGPT—from Python into Rust during the second quarter of 2026. The new Rust service is already handling 95% of OpenAI's production requests while using six times less CPU and 15 times less memory than its Python predecessor, with the company planning to deprecate the Python version entirely in the coming weeks. Back in March, Meta engineer Joe Savona led a port of the React Compiler from TypeScript to Rust that was "majority coded by AI," though its architecture, testing and migration strategy remained heavily human-directed.
Toub writes that AI agents wrote most of the code for GitHub's migration, noting that "a rewrite this size wasn't affordable before agents." He estimates the project would previously have required an entire team of developers working for one to two years, whereas it was "now completed primarily by a single developer, in only a few months." Sumner made a similar economic argument for Bun's rewrite, saying a conventional approach would have required a small engineering team for about a year—an investment that would have made the project unrealistic. "We could have kept fixing these kinds of bugs one-off in perpetuity, but we owe it to our users counting on us to do better than that, and systematically prevent these kinds of bugs from recurring," Sumner wrote at the time, explaining that Bun's growing reach to more than 22 million monthly CLI downloads had brought a persistent crop of memory-management bugs, including leaks and crashes.
The two companies arrived at similar conclusions about what coding agents make feasible, but their approaches differed sharply. Bun opted for an all-at-once port, using large numbers of Claude instances in parallel before bringing the Rust version back into the main codebase, with the Rust code passing Bun's existing tests on all six supported platforms after 11 days. GitHub chose a slower, incremental route, converting and shipping pieces of the Copilot runtime as they were completed across 128 pull requests while product development continued around the migration. According to the report, this incremental strategy allowed GitHub's team to catch and fix regressions along the way, though it stretched the timeline to more than three months compared to Bun's 11-day sprint.
A clear pattern is emerging: Rust was already drawing companies looking to rethink key software, and AI agents are now positioned as pivotal for migrations that may once have been too costly, disruptive or time-consuming to attempt. The report notes that for now, these various projects remain individual case studies, reported by the companies with the most to gain from demonstrating what coding agents can do on major software rewrites. Anthropic and GitHub are major vendors of the very coding agents they're crediting with making these projects possible, making their accounts inherently self-interested, though there are signs across the industry that the "agents do big Rust migration" phenomenon is gaining steam. The combined message from GitHub, Anthropic and OpenAI is that rewrites once dismissed as impractical are now within reach for small teams or even individual developers equipped with the right AI tooling. Yet the strategic calculus extends beyond simple productivity gains—companies betting on agent-assisted migrations are also signaling confidence that AI can maintain code quality and system stability at scale, a claim that will face scrutiny as these rewrites handle real-world production loads over time.