Linus Torvalds: What You Should Do As A Developer

Linux development is moving on a steady, reliability-first cadence—while open source’s broader ecosystem continues to expand in ways that make it easier for newcomers to build careers without gatekeeping. In a fireside discussion at the Linux Foundation’s Open Source Summit Europe in Vienna, Linus Torvalds emphasized that Linux’s most visible “updates” are often driver work rather than dramatic kernel rewrites, yet core kernel engineering still advances decades into the project’s life. He pointed to ongoing low-level work such as virtual file system code and long-running conversations around memory management, arguing that evolution is driven by both changing hardware and a growing user base.

Torvalds also framed release timing as intentionally unglamorous: Linux kernel releases land roughly every nine weeks. That rhythm, he said, is meant to keep changes timely and dependable rather than “exciting.” The discussion tied that approach to a major milestone: the completion of the real-time Linux project, described as taking about 20 years to reach fruition. Real-time Linux aims to add determinism to operating system behavior—an attribute that matters for workloads where timing guarantees are essential.

Beyond kernel mechanics, the conversation shifted to the open source landscape and how it’s being institutionalized. Announcements at the summit included the free 5GC open source 5G core software project moving under the Linux Foundation, Kamara issuing its first release to provide common network APIs for Telos and hyperscalers, and AWS moving its OpenSearch technology to new governance under the Linux Foundation. These moves signal that open source infrastructure is increasingly managed through formal foundations and shared governance rather than scattered communities.

Torvalds’ outlook on open source was notably optimistic despite industry layoffs and reduced corporate involvement. He argued that open source has become “granite” to many developers—something they expect to have access to—unlike earlier eras when even basic tools could require payment or effort to obtain. He linked that shift to the democratizing effect of open source: newcomers can use it as an entry point into the industry, build connections, and contribute even without the “right schools” or personal networks.

The discussion then turned into a practical debate about how developers should choose projects. Torvalds’ advice leaned toward finding something personally interesting that also matters to others, and avoiding the “lemming” chase for hype. But the conversation also included a counterpoint: building something purely for learning—or even something “useless to many people”—can still be career-shaping. The key test proposed was whether the builder enjoyed the work and whether it improved their skills. The takeaway was less about chasing stars or external validation and more about sustaining the internal motivation to build, learn, and get better—regardless of whether a crowd ever shows up.