I'm not a senior engineer by any means, despite all the time I've devoted to learning about CS and development in general; my main problem is that I just don't

I'm not a senior engineer by any means, despite all the time I've devoted to learning about CS and development in general; my main problem is that I just don't code anymore. That being said, I always have found that it is better to approach a problem with a model/plan in your head for how to implement your solution. It's fully possible to go in blind and just develop a solution iteratively, and that methodology can be useful sometimes, especially when working with something completely foreign. But if you can take the time to develop a model in your mind of how it works, then all that's left is translating the details of that model into code.