Wednesday, 30 September 2026 Colourful tech headlines from across the web

Agents Refactor 300K Lines in Three Weeks, and Practitioners Ask What It Proves

By Steef-Jan Wiggers · InfoQ · Wednesday, 30 September 2026 at 06:16 (16h ago) · Original article ↗

CodeScene has published a case study in which coding agents refactored 300,000 lines of C over three weeks for roughly $4,000 in tokens, verified by a frame-by-frame replay harness. The agents built a playbook of codebase-specific recipes along the way. Practitioners have questioned the scope, the metric, and how much of the result depends on the harness.

You're reading a preview. The full article is published by InfoQ on their website.

Read the full story on InfoQ

Google pulls back the curtain to tease Gemini 4 Argon

Google's next-gen Gemini is coming soon — but you can't try it just yet.

Android Authority 1h ago

Gemini for Home is now better at hearing you in noisy rooms

This is the third Google Home update of the month.

Android Authority 2h ago

Google announces Gemini 4 and says it’s so capable that only ‘trusted cyber defenders’ can have it right now

Google today revealed its next AI frontier model, which it's calling Gemini 4 Argon. The new model delivers "frontier performance in complex workflows across real-world software engineering, enterprise knowledge work like legal and finance, and cybersecurity defense," according…

The Verge 2h ago

Agents Refactor 300K Lines in Three Weeks, and Practitioners Ask What It Proves

More AI & Machine Learning

Google pulls back the curtain to tease Gemini 4 Argon

Gemini for Home is now better at hearing you in noisy rooms

Google announces Gemini 4 and says it’s so capable that only ‘trusted cyber defenders’ can have it right now