hey @bank-olad 👋 . welcome to github !
great that youre starting with data science as your focus ,,,a few practical tips specifically for building a portfolio that actually helps with employability :
on what to put in your repos :

• dont just upload raw jupyter notebooks with no context,,,every project repo should have a clear readme.md explaining what the project does,what dataset you used ,and what you found / built. recruiters /hiring managers skim readmes ,not code,when first looking at a profile,
• pick 2-3 solid projects rather than 20 half-finished ones,,,quality over quantityy matters a lot more for data science portfolios specifically.

good early project ideas that show real skills:

• an end-to-end project (data cleaning -> analysis -> visualization -> a simple model)using a real public dataset(kaggle is a great source),,,this shows the fulll pipeline, not just modeling
• something with clear visualizations ,,data science portfolios benefit a lot from visual storytelling(matplotlib /seaborn /plotly charts that actually explain something).

workflow habits worth building early:

• get comfortable with git itself(not just github the website),, commmit regularly with clear messages rather than one giant "final version" commit. this is something hiring managers genuinely check.
• pin your best 2-3 repos on your profile (customize your pins)so they're the first thing visible.

where to learn from here :

• search github for data science learning roadmaps /awesome-lists(search "awesome data science" ,,,theres a well-maintained curated list repo) ,,,tons of free structured resources linked there.

welcome aboard ,,,happy to help if you hit specific questions along the way💙💙💙💙 .