Data Science is a combination of Programming & Statistics, so to be a data scientist you need to have knowledge of at least one programming language, preferably Python/R as there is a good amount of people/communities who use these languages to build their models.
For a complete beginner, Python is easy to learn. Some of the basic tools used in data science from Python stack.
Jupyter Notebooks - IDE
Pandas - library for data manipulation and analysis
Numpy - library for scientific computing
Matplotlib & Seaborn - library for data visualization
Scikit-Learn - library for machine learning
Good mathematical knowledge helps to make a better judgment while choosing a procedure (algorithm) based on the data available to you and also to diagnose the problems.
If you don't have time to go through the theory, start with a tutorial. Follow the tutorial step-by-step. After you complete a tutorial, apply what you learned to new datasets. You can find some sample datasets online (https://www.kaggle.com/datasets). If you try the same modeling on a new dataset, you might run into a new issue. Upon doing some research, you might discover data issues in the dataset like different formats, or missing values.
This blog is first posted on hackerheap.com