Claude Code for Maintainable Software: Lessons from 4 Production Projects
See Also
The Sub-Agent Revolution: How Claude Code's Specialized Workflows Dramatically Increase Software Engineering Quality
Been reviewing some data from complex API development projects lately. Honestly, the traditional approach to AI-assisted development is pretty broken. We're asking general-purpose models to context-switch between domains like API design, contract testing, and security validation. The cognitive overhead alone kills productivity.
Then I Asked Whether Any of This Should Exist
By this point I had built the context-management layer three times — rented it, wrote my own, then built something more ambitious than either. So before building a fourth thing I stopped and ran the research: 99 agents, 78 claims, 25 adversarially verified. The answer was uncomfortable. The platforms are absorbing the layer.
The Three Meters Never Agree: Putting lean-ctx, rtk, and headroom on the Bench
Same real task, same harness, same models, one neutral meter — the provider's bill. A pre-registered protocol, right-of-reply issues filed with every maintainer, and every transcript public. Fourteen runs, ~76M tokens of my own money. What the bench said is not what any of the three meters promised.
More Articles
