Tapping into Data Science from a Non-Technical Background
This article is published as part of a series in our community (Women in AI & Robotics), sharing our diverse experience with everyone who feels like it’s time to take a step towards their career in STEM.
Why did you change to Data Science?
I had a dream. I was dreaming of a tool which automates part of my job as an Agile Coach: something that learns (and is able to predict) the good and bad patterns of communication and collaboration just by looking at conversational, unstructured data.
I figured the only way to find out if it’s doable was: to learn how to do it myself.
¯\_(ツ)_/¯
What kind of program did you do?
My rather unconventional path to Data Science started with building my own curriculum, self-teaching advanced mathematics and engineering basics. This meant hiring math and python tutors, working through Udacity’s Data Analytics and Data Science Nanodegrees and participating on two bootcamps:
- Ironhack’s project-based Data Analytics one and
- Data Science Retreat’s Machine & Deep Learning focused one
to develop a solid understanding of both the theory and practice of state-of-the-art models.
What kind of projects did you work on?
My first project which got more than 1k+ views on medium was a Sentiment Analysis based on our cohort’s public Slack conversations. To see how far I can go, I extended this one with a Social Network Analysis, using networkx’s library to identify the influencers of Slack communities.
Continued with tackling meetings that should have been an email —I teamed up with Jose (Mechanical Engineer) and Sam(PhD Neuroscience) to build our Meeting Notes AI, a tool which automatically transcribes your zoom meetings into text and then extracts a summary for you. You’re welcome. 😉
Was it hard to learn programming and the math?
Not going to lie, it was. Coming from a social science background, the first 6 months I spent eating chocolate to comfort myself between several emotional breakdowns caused either by an infinite amount of errors in my code or just generally looking at mathematical formulas.
But if in the end it makes you feel like this great lady here in a room full of men, I say it’s worth it.
Was it enough time to learn about Data Science?
Starting from zero, I got to a good base after a year, but since then I realised it’s a never-ending journey. My rather lonely side-projects helped me keep going, but I genuinely think there is no better way to expedite your learning than starting to work in a team as early as possible, getting to know how to deliver a solution from scratch and getting familiar with best practices around production-level code.
Would you recommend it?
Hell yes, it’s addictive. Not only is this the most exciting scene to be in right now, but learning more about data science will also put your personal life on the next level. For sure you’ll never look at visualised data in the news the same way again: distorted graphs, biased samples and misleading averages, here we go!
If this got you interested, go ahead and read Darrel Huff’s book: How to Lie with Statistics.
Do you work in Data Science now?
Yep, of course! My sweet spot is in the people/organisational development space, leveraging a mix of domain knowledge and product delivery experience from my background to create customer and business value through data.
This magic of mine will continue at CoachHub now, building the Data & Insights team’s roadmap and shaping the way data products serve the business as a Senior Data Product Manager.
