I'm a complete beginner and desperately need advice.

All I can tell you is be patient, stay at a slow pace, and just keep at it. I promise it will come, but it does take time and a lot of practice. I'm bad about posting my entire autobiography on here and spamming the same old stories, but I think sometimes what people need is one example of how you can take something just basic and spin it into something you use every day. The most condensed version is this: I got into Python after learning PureData for experimental and computer music. I wanted more automation and control, Python was easy to learn. PureData taught me a lot of computer programming concepts alongside MIDI automation and audio signal paths. And since that taught me about C-style expressions, the simplicity of Python was very attractive and it was a smooth transition from Pd to Python. Pd is *almost* a no-code environment, using visual objects you connect to get some result. But you can also easily write your own math or logic into custom objects (called "expressions). A Pd subpatch with custom expressions is virtually identical to Python functions. Next thing I know, I'm training TensorFlow models. I got a job at a library where I learned cataloging and ILS management, so I used Python for UI automation (macros), organized metadata in a SQLite database, and batch loaded new books into the ILS by the hundreds. I got so good at it I got fired. 🤣🤣🤣 Took a blue collar gig at a manufacturing plant, refused to sit during downtime, skipped breaks, reorganized/streamlined my station and the storage room, and made a mobile app to help me track production and estimate how much of what product I need to make. I started weighing in materials at the end of my shift to report usage to my boss while getting side-eyed by coworkers (nobody has ever done that before), and now he wants my work in writing. So now I'm writing a web app for warehouse management (WMS) and inventory. Basic functionality is done. Just need to add some features and move out of the experiment/testing phase. So my path has gone from part-time musician, public school teacher, librarian, cook, to cook/inventory manager and full stack developer, with the latter half of that last one possibly becoming a full-time job. I say all this not to brag. This is just me being me, which has gotten me in trouble more times than I care to admit, but it's also given me some pretty awesome moments, too. You're stuck for ideas, which is my whole point. Programming is about solving problems. If you don't have any ideas, it's because you don't have any problems to solve. My current boss came to me and asked "Do you know how to use Excel?" I couldn't stop laughing because I know as convenient as Excel can be, Python automation and SQLite databases will go further towards boosting productivity and efficiency. I had only one job when I walked into the production room for the first time. But it wasn't long before I saw all kinds of problems that could be easily solved with the right software. By doing all the development and deployment of a WMS, I'm saving my boss thousands of dollars. And I'm doing it in real-time over a network, not something easily done with Excel spreadsheets alone. Eventually I'll start turning in reports in Excel format, but even those can be automatically generated with very little effort. If I get moved to full time warehouse management, I can expand to cycle counts and give my boss access to the total warehouse and production landscape. Here's the thing: You said "And I always think 'someone already did this better'— which is funny, because I consider myself a creative person." Excel already exists and is really powerful. Well, Excel also integrates SQL. This is wonderful when my boss wants to see reports on materials consumption on the production floor. He wants to know where we're losing money or how much is being lost to waste/scrap, Excel is great. Do WMS software already exist? Of course. But they are super expensive. They also tend to be "kitchen sink" type apps with features you'll never use and shouldn't pay for, plus WMS software tends to be highly inflexible in how you use it. I'm working on a WMS that is tailored to how our plant operates along with our specific needs. When new features become necessary, I can simply add them as-needed. The wonderful thing about a Python/SQL WMS web app is its cross platform, can run on anything and anywhere. I could put it into a Raspberry Pi that attaches to a feeder. I can access it from my iPhone 12 to issue and return materials and clear out anything that's been consumed. Things that "someone already did better" might be perfect for Amazon or other large companies with extensive warehouse square footage, but for a small, family-owned plant, it might be overkill and too complex to learn and use effectively. Bigger isn't always better, and the "best" tools might not be the right tools for every situation. I have the power to scale my projects to my company's exact needs. So think about your exact situation. How can writing Python scripts or making web apps make your life and your work life easier, more accurate, and more efficient? How can you leverage your programming skill to outperform your coworkers, or how can you use your skills to create something that everyone can use to be more productive as a team? Start there, and I guarantee you're going to find plenty to do!