If your product's success depends solely on the models getting smarter, you're likely building the wrong thing. Instead, create something that improves as the models advance.

Building a new product often requires more than two months to determine if it meets customer needs; patience is essential.

I wouldn't expect AI models to one-shot entire products yet, especially if they're complicated. I prefer to implement features piece by piece and ensure each step works before proceeding.

When you're in the product market fit phase, it's crucial to keep refining your product instead of just focusing on scaling the business.

The better the AI models become, the more opportunities there are for startups to leverage these advancements. Startups need to be proactive in building their products in anticipation of new model releases rather than reacting to them.

If you have a breakthrough product, don't worry too much about its initial design; you can always improve it later with standard templates.

Developers should be agile and continuously experiment with technology, as many ideas may not work initially but could succeed in future iterations of the models.

You don't have to wait for the product to be perfect. Just launch it, develop it, and learn as you go.

A strong team in a great market will eventually build the right product, as the market will guide them towards it, even if it feels like a struggle.