'I'm not a programmer' anymore: Linus Torvalds on the only two tools he uses now

2026-07-08 · Source: News and Advice on the World's Latest Innovations | ZDNET · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning · Depth: Intermediate, medium

Summary

At Open Source Summit India 2026, Linux creator Linus Torvalds discussed the project's steady evolution and future direction. He highlighted Linux 7.1's incremental improvements, deliberately avoiding "splashy features." Torvalds, now primarily a "development lead" rather than a programmer, focuses on reviewing pull requests and understanding the "bigger picture," merging approximately 200 changes every two weeks. He expressed a lack of sentimentality for obsolete technology, noting Linux 7.2 will drop support for x86 machines without hardware floating point, such as the Intel 486 SX, released over 30 years ago. While acknowledging Rust's interest, Torvalds prefers C for its simplicity and robust verification tools like Sashiko, emphasizing that Rust doesn't prevent logic errors. He also addressed AI's role, noting that LLMs initially generated more "junk" than useful code, producing "hallucination" bug reports that consume significant human effort. However, he sees value in LLMs for prototyping and finding bugs, provided human developers validate and refine their output.

Key takeaway

For open-source project leads managing large, complex codebases, embrace a philosophy of steady, incremental improvements rather than chasing "splashy features." When integrating AI tools for code analysis or bug detection, ensure robust human validation processes are in place to filter "hallucination" and prevent resource drain from false positives. Your focus should remain on understanding the "bigger picture" of changes and fostering strong human collaboration, as even advanced languages or AI won't eliminate fundamental logic errors.

Key insights

Linus Torvalds advocates for steady, incremental kernel development, prioritizing human oversight and critical validation of AI-generated code and bug reports.

Principles

• Incremental development fosters stability over "splashy features".
• Hardware support must be balanced against maintenance burden.
• Language choice does not prevent fundamental logic errors.

Method

Torvalds operates as a development lead, reviewing pull requests for overall context and delegating code fixes, primarily using Git and email for communication.

In practice

• Validate AI-generated bug reports and code with human review.
• Prioritize "bigger picture" understanding in code reviews.
• Deprecate support for truly obsolete hardware to reduce burden.

Topics

• Linux Kernel
• Linus Torvalds
• Open-Source Development
• AI Code Analysis
• Hardware Support
• Git Workflow
• C vs Rust

Best for: Software Engineer, Machine Learning Engineer, Director of AI/ML

Related on AIssential

• Code Is Not Cheap: How to Multiply Your AI’s Output With Software Fundamentals — AI-assisted coding, while prevalent, demands human engineering expertise to mitigate quality issues from "jagged" model outputs.
• An Empirical Analysis of Static Analysis Methods for Detection and Mitigation of Code Library Hallucinations — Static analysis effectively detects code hallucinations, but inherent limitations prevent full resolution.
• MIT Prof. Explains AI-Assisted Programming: Part 2 — Combining LLMs with symbolic reasoning enhances code reliability and addresses ethical challenges in AI programming.
• From Hallucinations to Trust: A Human-in-the-Loop Playbook — Integrating human feedback is crucial for building trust and improving the accuracy of LLM-powered agents.
• not much happened today — Open-source AI models and agent harnesses are rapidly advancing, enabling powerful local inference and complex agentic workflows.
• Three more static code analysis sensors — Computational and inferential sensors significantly enhance AI coding agent self-correction and code maintainability, especially when combined with semantic interpretation.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by News and Advice on the World's Latest Innovations | ZDNET.