Boro: NVIDIA's Open-Source Effort For AI-Assisted Linux Kernel Development
([NVIDIA] 6 Hours Ago
NVIDIA Boro)
- Reference: 0001662684
- News link: https://www.phoronix.com/news/NVIDIA-Boro-Linux-Kernel-AI
- Source link:
Over the past few months NVIDIA has been developing Boro as a new AI-assisted kernel development CLI written in Rust and focused on local AI for enhancing efficiency for kernel development.
Andrea Righi of NVIDIA has been leading the development of this Boro AI-assisted tooling that is inspired by the successful [1]Sashiko AI tool out of Google and already doing [2]a great job providing code review on the mailing lists . Righi presented on Boro at this week's Linux Plumbers Conference (LPC 2026) in Prague.
Boro is focused on local AI driven patch validation for backporting of patches to older kernels, build/boot/testing of patches, and also plans for detecting possible out-of-tree module breakage. For those without the compute capacity for local AI, Boro also supports using remote OpenAI-compatible servers otherwise there are Claude, OpenCode, and Codex local back-ends.
While Boro is working on its own and can be used by kernel developers today, Righi expressed interest at LPC 2026 in potentially importing some of the functionality into Sashiko itself.
Boro is developed under an Apache 2.0 license. Those wishing to check it out can find Boro via [3]NVIDIA/boro on GitHub . Righi's LPC 2026 presentation can be found at [4]LPC.events .
[1] https://www.phoronix.com/search/Sashiko
[2] https://www.phoronix.com/news/Sashiko-Linux-AI-Metrics
[3] https://github.com/nvidia/boro
[4] https://lpc.events/event/20/contributions/2599/
Andrea Righi of NVIDIA has been leading the development of this Boro AI-assisted tooling that is inspired by the successful [1]Sashiko AI tool out of Google and already doing [2]a great job providing code review on the mailing lists . Righi presented on Boro at this week's Linux Plumbers Conference (LPC 2026) in Prague.
Boro is focused on local AI driven patch validation for backporting of patches to older kernels, build/boot/testing of patches, and also plans for detecting possible out-of-tree module breakage. For those without the compute capacity for local AI, Boro also supports using remote OpenAI-compatible servers otherwise there are Claude, OpenCode, and Codex local back-ends.
While Boro is working on its own and can be used by kernel developers today, Righi expressed interest at LPC 2026 in potentially importing some of the functionality into Sashiko itself.
Boro is developed under an Apache 2.0 license. Those wishing to check it out can find Boro via [3]NVIDIA/boro on GitHub . Righi's LPC 2026 presentation can be found at [4]LPC.events .
[1] https://www.phoronix.com/search/Sashiko
[2] https://www.phoronix.com/news/Sashiko-Linux-AI-Metrics
[3] https://github.com/nvidia/boro
[4] https://lpc.events/event/20/contributions/2599/