If I could wave a magic wand to become world class at any dev/product skill I'd pick "naming things". AI makes this a lot more attainable.
I have multiple 20+ turn chat sessions with Claude trying to come up with the right name for something: database schemas, file structures, public facing product/feature names, internal names, labels for UI artifacts, etc etc.
Being good at finding the right words to describe something is a super power. When you get it right it adds a ton of clarity and momentum as the team/customers/agents/etc can communicate about "the thing" using a shared vocabulary.
At this point Claude is soaked with rich context; just ask it to come up with 3 to 5 UI mock variations for you to consider “given the above context”. I promise the quality of those mocks and the inspiration you’ll get from them will blow your mind.
As a product builder, sometimes your highest leverage isn't building the product; it's building the tools to build "it".
Eval harnesses, internal admin panels, custom data viz.
There's never been a better time for internal-tooling.
I don't have a clean answer here. Something is clearly shifting. We do things at Huntr with 5 people that would've taken 30+ a few years ago.
But I think an explosion of new products and brutal consolidation within each niche can both be true at the same time.
If it costs millions in salaries to maintain a software product, you need a sizable addressable market to justify building it.
At $50K/year in tools, infra, and AI. Team of 1-3. A market of a few thousand paying users is a real business.
I think this changes what gets built.
If the minimum viable company is 1–2 people, we might be heading toward a world of solo operators and micro-partnerships more than small companies with employees.
"I work at a company" turns into "I am a company." Exciting if you're a builder. Less clear if you wanted stability