🚀 AI Is Changing What Matters in Software

AI can generate code faster than most teams can think. That does not mean the hard part is over. The software industry has spent years overvaluing output: more tickets closed, more features shipped, more code written, more visible motion mistaken for progress. AI supercharges all of that. It also reveals how incomplete that view always was. In enterprise software especially, the hardest part was never just making more things. It was deciding what matters, what is noise, what to preserve, what to cut, and how to keep systems and products coherent under real-world pressure.

AI is compressing the value of output and exposing the value of decision quality. It is exceptional at generation and acceleration, and increasingly useful in analysis and recommendation. But the highest-leverage work in software still lives in framing the problem well, setting the right constraints, making sound tradeoffs across technical and business realities, and staying accountable for the consequences over time. This talk argues that AI is not reducing the value of great engineers, product thinkers, or technical leaders. It is making their highest-value work easier to see. In a world of abundant generation, the differentiator is increasingly judgment.

• Development
• AI/ML