--- title: AMD and NVIDIA are both moving toward Rust in GPU software description: AMD is building a dedicated team to expand Rust usage across its GPU software stack, covering compilers, runtimes, firmware, and hardware/software co-design,... canonical: https://daily.dev/posts/amd-and-nvidia-are-both-moving-toward-rust-in-gpu-software-xanpfltki twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: AMD and NVIDIA are both moving toward Rust in GPU software | daily.dev og:description: AMD is building a dedicated team to expand Rust usage across its GPU software stack, covering compilers, runtimes, firmware, and hardware/software co-design,... og:url: https://daily.dev/posts/amd-and-nvidia-are-both-moving-toward-rust-in-gpu-software-xanpfltki og:image: https://api.daily.dev/og/posts/xanpFLTkI.png og:image:alt: AMD and NVIDIA are both moving toward Rust in GPU software og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # AMD and NVIDIA are both moving toward Rust in GPU software **[Collections](https://daily.dev/sources/collections)** · 3 min read · 17 upvotes · 0 comments ## Summary AMD is building a dedicated team to expand Rust usage across its GPU software stack, covering compilers, runtimes, firmware, and hardware/software co-design, and has an open engineering role for the effort. NVIDIA is also moving toward Rust, with its Nova Linux kernel driver already written in Rust and early-stage CUDA-focused libraries like cuda-oxide and Qile Rust in development. A smaller company, Vectorware, is building Rust GPU and Rust CUDA infrastructure to help standardize Rust tooling in a space long dominated by C++ and CUDA. None of these efforts are production-ready yet, and whether Rust becomes mainstream in GPU software depends on tooling maturity and real-world performance tradeoffs. ## Content ## AMD is building a dedicated Rust GPU team AMD Senior Fellow Harsh Meno recently announced the company is forming a small team specifically to push Rust deeper into its GPU software stack. The scope is broad: compilers, runtimes, firmware, and hardware/software co-design. An active job posting for a software development engineer confirms this isn't just exploratory - AMD is describing Rust as a core technical direction, covering performance-sensitive and high-trust parts of the stack. That's a meaningful signal. GPU software has been C++ territory for decades, and firmware especially tends to be conservative about language choices. Rust showing up there suggests AMD sees real value in the safety guarantees, not just the performance story. ## NVIDIA is moving in the same direction AMD isn't alone. NVIDIA has been pursuing its own Rust efforts, including the Nova Linux kernel driver and early-stage libraries like cuda-oxide and Qile Rust for CUDA-based GPU programming. The ecosystem around these is still immature, but a company called Vectorware is also working on Rust GPU and Rust CUDA tooling, trying to bring more standardized infrastructure to a space that's historically been dominated by C++ and CUDA with little else. None of this is production-ready at scale yet. But the direction is clear enough. ## NVIDIA also used AI to translate CUDA kernels into Rust On the more experimental end, NVIDIA built an agentic AI pipeline that translates CUDA Tile GPU kernels from cuTile Python and Triton-TileIR into cuTile Rust. cuTile Rust is a tile-based system that extends Rust's ownership model to GPU kernel authoring - an interesting idea, since ownership semantics map reasonably well onto the kinds of resource management problems that make GPU programming error-prone. The pipeline uses a multi-agent setup with specialized subagents handling analysis, kernel writing, host/FFI building, and performance validation. The team ran it against all 24 public TileGym operators (roughly 40 kernels total) and hit 99.5% of the original Python performance on average. Verification works by diffing the shared CUDA Tile IR between reference and translated kernels, which catches cases where a translation passes functional tests but gets the structure wrong. The tooling, coding rules, and converted kernels are all in the open-source TileGym repository. You'll need CUDA 13.1+, Rust 1.89+, and a Blackwell GPU to run the performance checks. ## What this actually means GPU programming has been stuck with C++ and CUDA for so long that any serious movement toward Rust is worth paying attention to. The combination of AMD building a dedicated team, NVIDIA experimenting with Rust kernels and drivers, and third parties trying to fill in the tooling gaps suggests this is more than a few engineers with a hobby project. That said, the GPU Rust ecosystem is still fragmented and early. The performance results from NVIDIA's translation work are encouraging, but translating existing kernels with an AI pipeline is different from writing new ones natively in Rust at scale. The real test will be whether the tooling matures enough that GPU programmers actually reach for Rust by default rather than as an experiment. ## Questions this post answers ### Is NVIDIA using Rust for its GPU drivers? Yes, NVIDIA's Nova Linux kernel driver is written in Rust. Early-stage libraries such as cuda-oxide and Qile Rust are also exploring Rust for CUDA-based GPU programming, though neither is production-ready yet. This reflects a broader industry move toward Rust for performance-critical, high-trust GPU systems code. _Engineers evaluating Rust for GPU work can follow driver and tooling progress like this on daily.dev._ ### Is AMD building Rust support into its GPU software stack? Yes, AMD has formed a dedicated team to push Rust deeper into its GPU software, targeting compilers, runtimes, firmware, and hardware/software co-design. AMD Senior Fellow Harsh Meno announced the effort, and AMD has an open job posting for a software engineer to help establish Rust across performance-sensitive and high-trust systems software. _Track how AMD's Rust GPU push develops before deciding whether to bet on it, via daily.dev._ ### What companies are building Rust alternatives to CUDA for GPU programming? Vectorware, a smaller company, is working on Rust GPU and Rust CUDA projects aimed at building more standardized infrastructure for GPU development in Rust. This is part of a wider effort, alongside AMD and NVIDIA initiatives, to close the tooling gap in an ecosystem long dominated by C++ and CUDA. _Developers weighing Rust against CUDA tooling can watch this space evolve on daily.dev._ ## Similar posts on daily.dev - [Introducing CUDA Rust: Two Tracks for Writing GPU Kernels](https://daily.dev/posts/introducing-cuda-rust-two-tracks-for-writing-gpu-kernels-a7096aegg) · NVIDIA Developer · 49 upvotes · 4 comments - [CUDA-Oxide 0.2 Brings Early Improvements To Pure Rust CUDA Kernels](https://daily.dev/posts/cuda-oxide-0-2-brings-early-improvements-to-pure-rust-cuda-kernels-glwfh6z2x) · Phoronix · 1 upvotes · 0 comments - [The future for Tyr](https://daily.dev/posts/the-future-for-tyr-xdzk3dk6f) · Hacker News · 1 upvotes · 0 comments --- Tags: [#rust](https://daily.dev/tags/rust), [#nvidia](https://daily.dev/tags/nvidia), [#gpu](https://daily.dev/tags/gpu), [#amd](https://daily.dev/tags/amd), [#cuda](https://daily.dev/tags/cuda) [View this post on daily.dev](https://daily.dev/posts/amd-and-nvidia-are-both-moving-toward-rust-in-gpu-software-xanpfltki) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"AMD and NVIDIA are both moving toward Rust in GPU software","url":"https://daily.dev/posts/amd-and-nvidia-are-both-moving-toward-rust-in-gpu-software-xanpfltki","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/amd-and-nvidia-are-both-moving-toward-rust-in-gpu-software-xanpfltki"},"datePublished":"2026-09-10T15:26:56.297Z","dateModified":"2026-09-16T17:05:10.713Z","description":"AMD is building a dedicated team to expand Rust usage across its GPU software stack, covering compilers, runtimes, firmware, and hardware/software co-design,...","image":"https://i.ytimg.com/vi/XqmOZ1TyLCk/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/XqmOZ1TyLCk/sddefault.jpg","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/amd-and-nvidia-are-both-moving-toward-rust-in-gpu-software-xanpfltki","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":17},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"rust,nvidia,gpu,amd,cuda","timeRequired":"PT3M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"AMD and NVIDIA are both moving toward Rust in GPU software"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/amd-and-nvidia-are-both-moving-toward-rust-in-gpu-software-xanpfltki#faq","mainEntity":[{"@type":"Question","name":"Is NVIDIA using Rust for its GPU drivers?","acceptedAnswer":{"@type":"Answer","text":"Yes, NVIDIA's Nova Linux kernel driver is written in Rust. Early-stage libraries such as cuda-oxide and Qile Rust are also exploring Rust for CUDA-based GPU programming, though neither is production-ready yet. This reflects a broader industry move toward Rust for performance-critical, high-trust GPU systems code. Engineers evaluating Rust for GPU work can follow driver and tooling progress like this on daily.dev."}},{"@type":"Question","name":"Is AMD building Rust support into its GPU software stack?","acceptedAnswer":{"@type":"Answer","text":"Yes, AMD has formed a dedicated team to push Rust deeper into its GPU software, targeting compilers, runtimes, firmware, and hardware/software co-design. AMD Senior Fellow Harsh Meno announced the effort, and AMD has an open job posting for a software engineer to help establish Rust across performance-sensitive and high-trust systems software. Track how AMD's Rust GPU push develops before deciding whether to bet on it, via daily.dev."}},{"@type":"Question","name":"What companies are building Rust alternatives to CUDA for GPU programming?","acceptedAnswer":{"@type":"Answer","text":"Vectorware, a smaller company, is working on Rust GPU and Rust CUDA projects aimed at building more standardized infrastructure for GPU development in Rust. This is part of a wider effort, alongside AMD and NVIDIA initiatives, to close the tooling gap in an ecosystem long dominated by C++ and CUDA. Developers weighing Rust against CUDA tooling can watch this space evolve on daily.dev."}}]} ```