From Engineer to Data Scientist: from Excel to Python
How learning new tools can change your career
Today I am writing about my career change from energy engineering to Data Science. A chance that naturally happened due to challenges that appeared in my engineering job.
Where did I start my journey?
Like many data scientists out there, I do not hold a degree in Computer Science. I am an Energy Engineer by background. Nonetheless, during my studies, I built a strong base for Data Science without noticing it. These are some of the silent strengths I built at university:
- Having a strong Math base: at university Algebra, Statistics, and Calculus courses have provided me with the necessary Math knowledge needed to thrive in Data Science. I never thought that such courses will be of great help to my career, but I now see that I was mistaken.
- Building different projects using Matlab: my faculty decided to bet on this advanced numerical tool and I used it for over 15 projects during my bachelor’s degree. It was a great start to get introduced to data analysis and programming.
- Enrolling in Intro to Computer Science and Web Development courses: despite these courses are not so relevant for data science, they helped me understand the whereabouts of programming. Those courses helped me getting a coding-mindset.
The first step into Data Science: Excel
After finishing my master’s studies I got a job in the manufacturing industry. Here I had to optimize the production process. The main target was to find the relationship between the process inputs with the end-product quality.
After learning about the production process, I came across some datasets that could be of help for my work, so that I started analyzing them one of the most used tools on Earth: Excel.
The datasets I was using were big enough for freezing Excel a few times per day and I was struggling to get good insights. This tool was blocking me, so I decided to start exploring other options for more advanced analytics such as Tableau or Microsoft PowerBI.
Getting insights from fancy visualizations: Microsoft PowerBI
Considering the limitations I had with Excel, I decided to invest some time with Microsoft PowerBI and the results were worth the effort!
This is a great tool for beautiful plots that loads big amounts of data with no performance issues on the PC. In addition, it tracks all the changes performed in the dataset and keeps the original dataset untouched.
With this tool, I obtained better insights mainly thanks to the powerful filters it has. Nevertheless, I still needed to go a step further for my work: I wanted a tool for making predictions.
My first language of Data Science: R
Having considered the limitations of the previous tools, I thought about getting my hands dirty and starting coding. Hence, I decided to further explore RStudio, the Integrated Development Environment (IDE) most used for R programming language.
Why did I start with R and not Python? The answer is simple: I already had RStudio installed on my work PC.
My serious data science journey began with the R for Data Science book, which mainly helped me in the following areas:
- Importing data
- Data cleaning
- Data preparation
- Data exploration
- Advanced Plots
With the new knowledge acquired, I prepared my datasets properly and made a wide variety of plots. It was a great data analysis exercise, but I was still missing the prediction part.
Making predictions with Machine Learning: Python
Machine Learning was a concept that I did not know until I spoke with a friend from my master’s. When I explained to him the challenges I was facing in my current position, he suggested me using Machine Learning.
I had nothing to lose, so I decided to invest some time in it and started a Machine Learning online course the following day. Here is where Python appears in the story. I could have taken the course in R, but I chose to go for Python to get a double benefit.
I had to invest a great number of hours in the course, but it was worth the sacrifice. My first predictions were finally made!
From engineer to Data Scientist: from Excel to Python
In my journey from Excel to Python, I naturally moved from the simplest tool to a more advanced one. Each tool gave me what I needed to advance to the next point.
Learning new tools made me take the following journey:
- Engineer: first analysis with Excel.
- Data Analyst: more advanced analysis with Microsoft PowerBI.
- Data Scientist: first, I started cleaning, preparing, and thoroughly exploring the data with R and RStudio. Then, I broke into Machine Learning using Python.
After going through that steep learning curve, I changed to another firm to work as a full-time Data Scientist. This proves that the right experience and attitude can help us to achieve a job in Data Science no matter what background we have.
