Slide 1

Slide 1 text

Working as a Data Scientist Elvyna Tunggawan

Slide 2

Slide 2 text

A few things about me ● Took an undergraduate degree in Information System ○ Graduated from Universitas Multimedia Nusantara, Serpong in 2016 ● Worked as (Associate) Data Scientist in Airy Indonesia (June 2016 - Feb 2020) ● Currently a master student at The University of Auckland, New Zealand

Slide 3

Slide 3 text

How could I become a data scientist? Disclaimer It was not a common career path for an IS graduate Not to mention that “data scientist” was not a buzzword yet!

Slide 4

Slide 4 text

What I knew back then ● Database systems ● Programming ● (A bit) statistics ● (A bit) data mining ● Data warehousing ● Business intelligence

Slide 5

Slide 5 text

How does the daily work of a data scientist look like?

Slide 6

Slide 6 text

Source: R-bloggers

Slide 7

Slide 7 text

Source: Analytics India Mag

Slide 8

Slide 8 text

Source: Microsoft - TDSP

Slide 9

Slide 9 text

More importantly, data scientists transform data into business values.

Slide 10

Slide 10 text

Source: Reddit

Slide 11

Slide 11 text

Sample work Source: Airy ♥ Science - Medium

Slide 12

Slide 12 text

What skills should I prepare?

Slide 13

Slide 13 text

Familiar with this venn diagram? Source: Towards Data Science

Slide 14

Slide 14 text

Structured Query Languages ● Window function ● Query optimization

Slide 15

Slide 15 text

Programming Familiarity with Linux environment is a plus!

Slide 16

Slide 16 text

Data Visualization What can you tell from these pie charts? Much better, right? Source: The Next Web

Slide 17

Slide 17 text

Visualization can be misleading Source: The Next Web

Slide 18

Slide 18 text

Data Visualization Tools

Slide 19

Slide 19 text

Statistics If you have to choose a single number to represent the distribution below, what summary statistic will you choose? “Average” could refer to different things! Is it mean, median, or mode?

Slide 20

Slide 20 text

We can take wrong decisions without understanding stats

Slide 21

Slide 21 text

Machine learning Source: xkcd ● Supervised learning ● Unsupervised learning ● Semi-supervised learning ● Reinforcement learning

Slide 22

Slide 22 text

We need more than technical skills! Curiosity Critical thinking Growth mindset Communication skill

Slide 23

Slide 23 text

Any challenges?

Slide 24

Slide 24 text

Data quality!

Slide 25

Slide 25 text

Source: Forbes

Slide 26

Slide 26 text

People don’t really know what they need. They ask questions, but not the right one. Source: reinerbotha.com

Slide 27

Slide 27 text

Uncertainties - business evolves quickly!

Slide 28

Slide 28 text

Communication ● Explain the statistical and math terms in an interpretable way ● Business people do not really care about the technical stuffs you do!

Slide 29

Slide 29 text

What are the career options?

Slide 30

Slide 30 text

Industry vs. Academia vs. Freelance ● Research ● Teaching ● Service ● Check upwork.com / freelancer.com ● Make money from data science competitions ● Be an entrepreneur ○ I’d suggest you have a prior industry experience ● Data scientist ● Data engineer ● Machine learning engineer ● Data analyst ● Business intelligence analyst ● Statistician

Slide 31

Slide 31 text

Thank you! Reach me on Twitter: @vexenta