The Simple Process To Get Real World Data Science Experience
Do you feel like you do not stack up well against other people looking for a data science job because you have no "real world experience"?
You have a good combination of academic background and non-traditional training (Coursera, Udacity, and several other MOOC's) under your belt. Yet, when you look at job requirements for data science jobs, it causes you to get discouraged because you don't feel like you don't have enough or even any "real world experience". Doing a data science job search is draining because it feels like a catch-22 in that you need experience to get hired and the only way to get experience is to already be working as a data scientist.
Rather than wait for a job to give you "real world experience", start small and develop it yourself
The way to obliterate a catch-22 situation is to sneak up on it from the side. Rather than wait for a data science job to give you "real world experience", you can start with small data science projects which will build your experience in using data science to solve problems, answer questions, and build insights. So at a very high level, the process is a) find a data set you are interested, b) ask and try to answer questions about it, c) write up the results, and d) rinse and repeat. Doing this process will give you all of the "real world experience" you need in order to get a data science job.
"Real world experience" is something to work towards, it is not something you receive
There is no on/off switch that you can flip that will give you the label of having "real world experience". It is something that you gradually gain as you work with different libraries, frameworks, and tools. It is something that you work towards as you download, work with, and collect different data sets. Working with small, "Grande", large, huge, and internet scale data sets will teach you something different for each size. Same thing with working with very clean data sets and very messy data sets. Having "real world experience" just means that you have done data science and learned by immersion.
"Real world experience" is used to measure how effective you are as a data scientist
"Real world" is used as the opposite of "academic". Or said another way, "real world" means that you've had to deal with all the issues of doing data science, not just using clean data to produce a model. It means that you understand the business problem behind the question being asked or the business function that is generating the data. It means that you have chosen what metric(s) to measure and test. It means you have decided what data to use. It means that you have cleaned up your data. It means that you chosen a model to use and actually built it. It means you measured your models performance.
Doing all of those steps and actually thinking through them several times will make you more effective the next time you work through a project. You will have some thoughts and ideas of which approaches to use, which tools to use, what is important and what is less important. All of this knowledge will make you a more effective data scientist. It's important to remember though, that it's not an off/on switch where once you've done a few projects you can label yourself "effective". It's an ongoing process.
There is nothing magical about "real world experience" in data science
If you look at the curriculum of most "data science bootcamps" and "data science fellowships", you'll find that they are almost all project based. This is because there is nothing magical about developing enough "real world experience" to get a data science job. The development is based on doing projects that differ in data, tools, scope, size, software, statistics, and math to give the participants a feel for what it's like to work as a data scientist. Which means you can get a similar experience in developing the understanding of the concepts and paradigms that real data sciences encounter by working through projects.
Make yourself more desirable to a hiring manager by developing "real world experience" in data science by working through lots of small projects
Don't get discouraged because of the catch-22 of needing "real world experience" to get a data science job. Follow the process of a) finding a data set you are interested, b) asking and trying to answer questions about it, c) write up the results, and d) rinse and repeat to develop "real world experience". Doing this process will give you all of the "real world experience" you need in order to get a data science job. So the next time you see a data science job posting and it asks for "real world experience" think about what small project you could do with a data set related to the job posting that will allow you develop effective data science skills and experience. Good luck!