A paradigm shift in how software is built
There is a paradigm shift in how software is built. One multidisciplinary person — with a product mindset, an engineering understanding, and good taste — can now build products alone, using tools like Cursor and Claude, at a fraction of the time and cost.
The traditional need for separate designers, product managers, developers, and QA is starting to blur. Much of the process can now be handled through agents, combined with deep thinking about how the product should work — both at a fundamental level and in its UI/UX. The company of one is now a possibility.
Over the past few weeks, I've been experimenting with these tools. I have never felt more productive. From building simple API wrappers to fully-fledged products in days instead of months — this is already reality. In many cases, it feels like a 100x increase in productivity. It feels like being a maestro guiding an orchestra. This is a tectonic shift in how software is built. And agents will only become more capable.
Right now, there is excitement — the feeling of amplified productivity. But what lies ahead? How will this change our industry? What happens when even the “maestro” is no longer needed? How will this new capability be used in practice? And what happens when these systems extend beyond software — when they start operating in the physical world, through robots and automation?
One thing feels certain: we are living through a pivotal moment in human history. We may look back at this period as the moment when a form of singularity arrived for software — when the act of building fundamentally changed.