My advice is not to focus on a portfolio or a "project". Find things that annoy you in life that code can fix, write that code. Some of these things will become more complex then others. I would advise against anything larger than 3 days of work (as in distinct days spent working on the project ,not total hours) in scale, as you'll just feel bad when you don't finish the latest grandiose dream.
Don't worry about how pretty or impressive the code is. The point is the function. Save everything in a projects folder, no matter how shit.
At some point you'll hit a point where a project hits the right intersection of elegance/function/polish, and then you can throw that on a portfolio page. Then another one, and another one.
The important thing is the end goal. As long as that goal is to impress other people you will be paralyzed by fear of rejection. If the goal is to fix a problem for yourself that's much more achievable, as the only bar you need to meet is "did it fix my problem?".
Random sampling of my own "frustration" projects:
- automatic version control in SQL Server for object/table definitions
- cron script that scrapes + downloads the latest driver from Nvidia from their website, installs it, then deletes the file
- command line tool that formats text in the clipboard various ways (mostly I use this for transforming csv/excel files into SQL lists or table inserts)
- a small utility for printing notes on a receipt thermal printer
- a Lemmy scraper that combines different communities into a single coherent image board (I hated their default media presentation)
Don't worry about how pretty or impressive the code is. The point is the function. Save everything in a projects folder, no matter how shit.
At some point you'll hit a point where a project hits the right intersection of elegance/function/polish, and then you can throw that on a portfolio page. Then another one, and another one.
The important thing is the end goal. As long as that goal is to impress other people you will be paralyzed by fear of rejection. If the goal is to fix a problem for yourself that's much more achievable, as the only bar you need to meet is "did it fix my problem?".
Random sampling of my own "frustration" projects: - automatic version control in SQL Server for object/table definitions - cron script that scrapes + downloads the latest driver from Nvidia from their website, installs it, then deletes the file - command line tool that formats text in the clipboard various ways (mostly I use this for transforming csv/excel files into SQL lists or table inserts) - a small utility for printing notes on a receipt thermal printer - a Lemmy scraper that combines different communities into a single coherent image board (I hated their default media presentation)