Learning to Swim vs. Taking the Speedboat (i will not promote)

Hey everyone! Although I've been involved in software development for a while, I wouldn't even consider myself a junior developer, since I've never had a proper job in the field (apart from a paid, part-time university project). Over the past few months, I've been developing a web app and a mobile app. They're two different products, but they're complementary to each other. Along the way, I've been learning the fundamentals of Flutter, FastAPI, and SQL, as well as software development concepts, particularly those related to mobile development. At this point, I'd estimate the project is around 60–70% complete. However, I've built almost everything the old-fashioned way: learning as I go and writing the code myself. My use of AI has been limited to free plans and web-based interfaces. I mainly use AI as a learning tool. I spend a lot of time discussing things with it, and whenever it generates code, I make sure to read, understand, and review everything. At the end of the day, everything goes through my hands. My original plan was to start relying much more heavily on AI after finishing the MVP. However, given how far AI has come, I'm wondering whether I should change my approach now rather than later. On the one hand, I could keep developing the way I have been so far, strengthen my fundamentals, and maintain control over the codebase. On the other hand, I could start using AI tools much more intensively, delegate a significant portion of the implementation, and focus on finishing the product and launching it as soon as possible. I estimate that completing the remaining 30–40% would take me around four months with my current approach. By leveraging AI much more heavily, I might be able to cut that down to just one month. My concern is that by accelerating development this way, I might lose some control over the codebase and miss out on learning fundamentals that I still need to master. What would you do in my position? Would you stick to a more traditional approach until the MVP is finished, or start working much more intensively with AI right now? How would you strike a balance between learning to code properly and taking advantage of tools that can dramatically speed up development?