A Minimalist in a Big Data World
Just over a year of studying (online) and I’m applying for Data Science jobs!
Last year during my career break I decided to focus my study towards Data Science (as the web development and computer programming side of “coding” did not interest me as much as Data!) and now I feel I have acquired all the skills and tools needed to get a Data Science job. So last week I started applying for internships and entry-level jobs to gain some real-world experience. I applied for 3 internships and 4 entry-level jobs and just 3 days later one company has already rejected me and sent me an (automatic) reply:
“we regret to inform you we will be moving forward with other candidates who more closely match the requirements of the position.”
Downtrodden and lost I decided to blog again, to let off steam but also to put thoughts into words in order to clear my mind, remain focused and mainly to document to potential employers that I am the candidate who can closely match the requirements of the position.
In the back of my mind I have always known that a career change from Civil Engineering to Data Science was not going to be “all sunshine and rainbows” (to borrow a quote from the motivational Rocky Balboa!), and that’s why I am aiming to start from the bottom, going for internships and entry levels jobs, I am not afraid of starting all over, down but not out!
Further on in the rejection email, they kindly suggested:
“We encourage you to continue searching for job opportunities on our career site and applying to any new positions that match your skills and interests.”
The latter, ‘interests’, is not an issue as to make an enormous jump from Civil Engineering to Data Science would be crazy unless I wasn’t a tiny bit interested in Data Science (that’s that one ticked off the list!).
Ok, the ‘skills’ part, how do I demonstrate to a potential employer I have the skills when:
- I have not studied or not in the pursuit of a Computer Science, Data Science or Statistics degree… or even modeling-heavy discipline (e.g. economics, econometrics, linguistics, neuroscience, astrophysics, atmospheric science, geophysics, etc.)
- Do not have 1 or more years experience as a data scientist
- Cannot demonstrate my Data Science skills on paper (or .doc / .pdf formats in this case) without a chance of an interview.
I plan to address the above 3 problems in the following way (problem-solving, that’s a great skill to have isn’t it!):
- All I can do here is note the online courses I have taken, and from my perspective and experience in completing (certificates as proof!) these online, I would highly recommend the following courses (in no preferential order):
The Python Mega Course: Build 10 Real World Applications by Ardit Sulce
Normally you find a free course teaching you the basics of a programming language and in this case Python, nothing wrong with that but with this course, it takes it a step further. After you learn a new technique or a new syntax, Ardit instantly teaches you to build an actual “real-world” application such as Webmaps, Webcam Motion Detector and many more (8 more to be precise!). This reinforces your learning no end, you learn then you create.
Ardit’s delivery and hands-on approach make learning Python for beginners such a less intimidating task, there are lots of resources in this course and Ardit answers any questions himself and the course is updated regularly!
Python for Data Science and Machine Learning Bootcamp by Jose Portilla
Just like Ardit, Jose is a great lecturer, clear and concise explanations in the video lectures, backed up by enormous amounts of resources including the code in the videos (course notes), and homework with solutions. Jose is backed up by his Pierian Data team (teaching assistants) who help answer questions but Jose does chip in to help out!
This course takes you through a crash course in Python basics then leads you into using Python and Pandas for Data Analysis before heading into Data Visualization (he details Matplotlib, Seaborn, Plotly and Cufflinks), I fell in love with Plotly so much so I took another of Jose’s courses, Interactive Python Dashboards with Plotly and Dash to learn yet another skill. It is perfect for learning, each lecture builds seamlessly upon the last and the course finally dives into Machine Learning.
The Machine Learning section is exhaustive, Jose and the team go through all the classic Machine Learning techniques from Linear Regression to Logistic Regression with Sci-Kit Learn and they also teach Random Forests, K Means Clustering and even Natural Language Processing (all in detail). There is also a section on Spark and Deep Learning with TensorFlow, it is a jammed packed course and well worth any amount of money (I got this when it was on sale for $9.99 — bargain!).
As you will note I prefer learning on Udemy and that is a personal preference because I could pick and choose courses specifically to fill my skill gaps and for a very reasonable price, so far I have self-taught myself Data Science for less than $80! But here I tried a different platform for comparison, LinkedIn Learning.
You can get a month free trial, and that is what I did and in that month I took “Learning Python” by Joe Marini, I deliberately took this course to solidify my basic Python knowledge and to test myself but also because the instructor uses a different IDE (Microsoft’s Visual Studio Code) and that taught me to use another environment to code, so remember not to limit yourself.
Right, back to reviewing Joe’s course, nothing too different from most basic Python courses and that’s a good thing, Joe’s delivery is understandable and he explains his thought process perfectly, which is great as in learning to code or in Data Science it is important to know how and why things are done in a certain manner.
There are exercise files included for download and what is good is it comes with a “start” file and a “finish” file so if you struggle to get the same outcome as the instructor you can always check his code!
Data Science A-Z: Real Life Data Science Exercises Included by Kirill Eremenko
As I progressed further into my self-study I noticed that whilst it was great to learn Python, Pandas, Sci-kit Learn, Matplotlib and PCA but how do you combine all this in the real-world or indeed the Data Science world. Because of the courses I took and my interests, Udemy’s algorithms (I’m sure Machine Learning profiled me) suggested Kirill’s course and what an ideal course it was to show me the Data Science world.
This is exactly what Kirill does and with enthusiasm, he describes the whole Data Science Lifecycle in four distinct sections, Data Prep -> Visualization for Data Prep -> Modelling -> Communication & Presentation. It is through this course that I knew Data Science is perfect for me and he introduced me to ETL (Extract — Transform — Load) using Microsoft tools, Excel and SQL Server Integration Services (SSIS), what a great skill to have.
What stood out most for me on this course and the other courses I took on Udemy is that the instructors are actual professionals (or Data Scientists) themselves and it shows, they not only pass on their knowledge but also their real-life experiences with their preferred tools (as in all of programming there is no one solution to a problem!).
I thoroughly enjoyed his style of delivery (it’s almost like he didn’t edit out his mistakes in the video to purposefully show you how to correct them) and how he is a keen advocate of keeping a trail of quality control and assurance so you can track and correct your errors with ease. Again this course was well worth my $9.99 (Black Friday sale!).
The Data Science Course 2019: Complete Data Science Bootcamp
I have currently completed 98% of this exhaustive course, there are 21 hours of video lectures alone! This course delves much deeper into Data Science from its evolution to applying all the techniques taught in a business case study about “absenteeism”. Not to favor any of the courses I have reviewed above so far but if you were to choose only one course on Data Science this one would fulfill them all. As I said it is exhaustive, it provides all the skills you need as a Data Scientist: Mathematics, Statistics, Python, Advanced Statistics in Python, Machine & Deep Learning!
Again, don’t be daunted by all the content, if you find some of the topics too basic you can skip it, but I followed through every video lecture and exercise so far and it has taught me so much and refreshed my high school statistics that I thought I would never ever use!
You can’t go wrong with this course, so I would recommend anyone struggling to learn Data Science online that not only teaches the theory but also teaches application should go for this course at only $11.99. The 365 Careers team have done a great job and the teaching assistants answered all my questions within hours not days!
I feel it is this course that will help me get a Data Science job one day, it made me understand why 70–80% of a Data Science project is data cleaning and they do spend a lot of quality time on elaborating on the preprocessing process.
2. This one I cannot show or prove, I would be lying if I said I had one or more years working as a Data Scientist or similar role. I will just have to hope and pray that someone will eventually consider me for an internship or entry-level job and I can start to gain the ‘experience’ I require to continually learn and develop.
3. I’ve decided since I cannot actually showcase my skills on a piece of paper (resume or cover letter) I shall instead blog here on a weekly basis, each week discussing and demonstrating a skill relating to each stage of a typical Data Science Lifecycle (as shown below). There are some neat tips and tricks I would like to share as well, which have saved me time and increased my productivity when learning Data Science skills.
As I have linked to my medium blog on my resume and cover letter, I hope any potential employer will take note of my upcoming blogs demonstrating my data science skills, my communication skills (I created that Data Science Lifecycle infographic myself on Microsoft PowerPoint!) and can get a sense of my keen interest/passion for data science (if not a little desperate).
On a side note, is this minimalism, I seem to be doing/creating a lot? I guess what I am doing is minimalism, here I am focusing on what is of value to me and decluttering because I am not concerned with my past career (which wasn’t working out for me (not in the way I had hoped anyway).
Please get in touch if you have faced a similar situation when trying to get your first Data Science job and if you have any advice that would be awesome too!
