As our CFO @_balaji_km mentioned at earnings today, we’re seeing some very interesting trends on AI costs. I think it’s another signal that we’re coming to the end of the so-called ‘tokenmaxxing’ era.
Here’s what’s been happening behind the scenes.
Since the beginning of the
Excited about the future of intelligent software development. CEO & co-founder @QodoAI
- 1/ the harness isn't scaffolding you delete for every model update. It should be a meaningful part of the product. If you can just delete it and get better results, it was never a moat.No idea if these specific numbers generalize across tasks, but directionally it’s clear that the harness is going to become the most important variable -right next to model capability- in the AI stack. The ability for harnesses to break down work in the most efficient way and
- In a review of our cybersecurity evaluations, we found three incidents in which a Claude model reached the internet from within or while interacting with a third-party evaluation environment, and then gained unauthorized access to the real systems of three different
- @unclebobmartin doesn't read the code his agents write. @dexhorthy says, for now, read the dang code. It kicked off some great discussion and follow-ups, e.g. @GergelyOrosz, because they're both right. Wrote a piece about it.
ArticleThe Code Is Disposable. The Codebase Is Not.Uncle Bob recently answered an old software engineering interview question in a very 2026 way. The question: Imagine you lose your entire software system, but you can recover one of two disks. One...







