How to Grow As Data Scientists
Have you felt that you already became a super data scientist when you see an advertisement about a data scientist course saying you can become one π¦ΈββοΈ in 3 months?
Have you lost motivation and purpose of working other than making π° it when your daily routine is nothing but reading and copy-pasting your or others' work?
If so, you might think something differently.
Here are two tracks:
The Fixed Mindset
On the contrary, it is:
The Growth Mindset
If we take a look at these two tracks:
Iβm sure you got the idea of which is β or βοΈ. And more importantly, learning the growth mindset is so easy π° or even just π§ that everybody can learn and grasp.
However, if you ask me now that I get the growth mindset, what are the best practices I need to do then?
Let the Kaggle Survey 2020 tells us these three questionsβ¦
- How to grow as data scientists?
- What should data scientists think important in 2020?
- When you want a higher annual compensation, what are the considerations?
Here are the answers I found through my Kaggle Survey 2020 analysis π΅οΈββοΈ
1. Growth πππ
- Focus on learning new things in the first 4 years of your career. Find your job needs, build up, and stick to 4β5 skills as your core π€πΌπ€πΌπ€πΌ to make yourself unique and stand out. At the same time, as data scientists, donβt forget to be open-minded and have some knowledge of other things (after all, we take multiple job responsibilities instead of a single one)
- Engage with the community and interact with others by sharing your work and learning from others, it is all-time important!
2. Whatβs Important to Know βββ
- Having at least a Masterβs degree is essential for data-related jobs, but itβs becoming less important
- Know Python π, learn π, master π, and integrate π with the other languages you regularly use to build a network πΈοΈ in your brain π§
- Free services like Google Colab and Kaggle Notebook are the must to know but donβt completely ignore pay services like Amazon Sagemaker because pay π° could mean better experiences and better quality, and it might be closer to the industry needs (look at AWS Sagemaker, it has a huge gap between the experienced and students)
- Make visualizations aesthetical, easy-to-use, and interactive. Data Scientists have the β1 usage in Seaborn, Ggplot, Plotly, Shiny, Bokeh, Leaflet/Folium comparing with other data related people. Especially for Plotly β¨, comparing with other jobs, data scientists have more than 10% of people using it.
3. When You Consider A Higher Annual Compensation βββ
- Career experience is the most important aspect: one needs to know how to drive large business impacts using data and their skills. It also needs to know how to build things on large scale by knowing what different scales of companies or data teams are doing πΆ-π§-π§πΌ.
- Skillsets and abilities are the 2nd important aspect: Plotly is strongly recommended! Or any other interactive package also works. And donβt forget to know how to put it in production π±. Besides, we shall always get to know and apply more IDEs, languages, visualizations, and machine learning frameworks whenever needed. p.s. If you are having a data job in the United States, youβve been on the pirate boat β-π°.
Well, a growth mindset makes you as πππ and keep you from ποΈποΈποΈ. And you can wire it up in just 1 second! ππΌππΌππΌ
Last, if you are interested in learning more about other details in the analysis, Check out my notebook π!