What are some of the challenges of becoming a data scientist?
I am a student who has passion for STEM. I am struggling to figure out which career path I belong to. I want a profession that is not boring and something I will love. I think I want to become a data scientist. #stem #technology
For sure there are some challenges that exist with becoming a data scientist BUT before I share some, I hope you know that many of the challenges can be overcome...To be clear, my professional title is not as a data scientist but the work I do can be classified as data science work. In general terms, data science includes using statistical methods and tools, including AI and data visualizations, that help to provide insight on how to make something better. For example how to do something faster, how to get more users on an app, how to get more sales for a product. Because data science can be used in so many places and for so many reasons, one of the challenges can include finding the right data to solve the problem. Being able to understand the data, sometimes in it's rawest form, can also be a challenge. In short if the raw data isn't "clean" (that is if it isn't meaningful, useful, accurate), then any conclusions that come from it won't be helpful to meeting the goal(s). A strong background in math an statistics is also helpful in this area, though I think this more of an initial challenge that can be overcome.
Knowing what tools to use to help get to what can solve the problem can also be a challenge. Generally speaking data science tools and methods become important when you are working with extremely large sets of data...like too big to open in a spreadsheet :). In these cases learning how to use tools that can ingest large data sets and tell important stories about what matters in that data, trends that either need to improve or need to change...learning how to use these tools can be an initial challenge but can be overcome.
One last item that can be a challenge (that can be overcome :), is communicating findings in a way that isn't overly techy. Specifically, data scientists can be wildly successful in seeing problems according to the data. Sometimes though, the data scientist isn't directly involved in the mechanics of fixing a problem. Instead they have to communicate the problem to "the fixers". Being able to communicate what the problem is in a way that helps "the fixers" execute can end up being a win-win for all team members involved.
Hope you find this response helpful. Best of luck to you!
What is interesting about being a Data Scientist, is the combination of skills from what it takes to work with 'Data' and drive analytical insights as a 'Scientist'. If you enjoy working with numbers i.e. mathematics, solving challenging problems and/or have an interest in computer programming these are excellent interest areas that combined enable Data Scientists to do their roles.
Note: You do not have to have all three to get into a Data Analyst role, but the combination of working with data, understanding mathematical models, and programming key algorithms combined can be a very powerful combination to succeed.
One item I always stress, is ensuring before looking to 'solve' or come up with a solution, is to ensure asking the right set of questions that needs to be solved to ensure the right set of data and insights are applied.
For example, recently I had worked with my team and we were researching why customers were not using a recently released product as much as expected. The original 'Problem' statement the team started to look at was to determine what capabilities/features the product didn't have which prevented it's usage. However after doing some initial research and surveying certain customers, the real problem statement that became apparent was customers were not aware of the product and this allowed the team to shift to research how best to market/drive awareness of the Product vs Product features/capabilities.
You are a very lucky to be in the best period to study and work in data science.
I graduated in Europe, with a Master in AI and vision. Problem: that was 25 years ago and computers able to handle AI were hard to find, expensive, use cases impossible to solve with the algorithm we knew by then and therefore, jobs were not available.
But today, this is not the case anymore. Technology is there, jobs are there, and we have problems solved every day using AI.
Now AI won't work without data. And more importantly, unbiased data. That is where the Data Scientist enters the loop. As it was mentioned above in another reply, dealing with data is a skill. You'll have to understand what data you are working with, how they are correlated/uncorrelated, biased/unbiased, what data to add into your algorithm to enhance the performance, analyzing a performance, understanding what to tune in order to improve the performance. With experience, you will work on more complex subjects, requiring thousands if not millions of parameters.
So if you are up to going into details when something doesn't work, but also taking a bird eye view to consider an overall situation, understand what customers (internal or external to your company) are asking you to solve or improve, with a solid STEM background, you will do great.
I would recommend you don't stop too early your studies and aim for a Master degree at minimum so 2 years you will invest in your early time, will save you maybe 5 or 10 years in your career progression.
All the best!