Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Build a Career in Data Science

Emily Robinson
November 10, 2019

Build a Career in Data Science

“The best job in America.” “The sexist job of the 21st century.” Data Scientist, a title that didn’t even exist before 2008, is now the position employers can’t hire enough of and job seekers strive to become. With this popularity comes more and more people vying for entry-level data science jobs. How can you stand out from the crowd and actually land your first job as a data scientist?

In this talk, Emily will cover what skills you need to start your career, the different types of data scientist jobs, and how to best position yourself based on your academic and work history. She’ll show you how to make not only a standout resume and cover letter but also a strong data science portfolio of projects and blog posts. If you’ve been struggling to break into the field or are even just curious about what the data science hype is all about, this talk is for you.

Emily Robinson

November 10, 2019
Tweet

More Decks by Emily Robinson

Other Decks in Business

Transcript

  1. Build a Career in
    Data Science
    Emily Robinson
    @robinson_es

    View Slide

  2. About Me
    ➔ Data Scientist for 3 years
    ➔ Background in statistics & social sciences
    ➔ Writing “Build a Career in Data Science”
    with Jacqueline Nolis

    View Slide

  3. Build a Career in Data Science
    9 chapters out now, rest early 2020
    40% off with code mtpdcds19 at datascicareer.com
    (code is good for 40% off everything on Manning)
    Build a Career in Data Science teaches you
    what school leaves out, from how to land you
    first job, to the lifecycle of a data science
    project, and even how to become a manager.

    View Slide

  4. Build a Career in Data Science

    View Slide

  5. What is Data Science?

    View Slide

  6. One definition
    https://hackernoon.com/what-on-earth-is-data-science-eb1237d8cb37, Cassie Kozyrkov
    Data science is the
    discipline of making
    data useful

    View Slide

  7. Classic data science venn diagram
    http://drewconway.com/zia/2013/3/26/the-data-science-venn-diagram

    View Slide

  8. Our (slightly updated) version

    View Slide

  9. Programming: What you need to know
    OR

    View Slide

  10. Mathematics & statistics
    1. What techniques exists
    • I need to group customers together -> I should try clustering
    2. How to apply them
    • How to do a k-means clustering in R/Python
    3. How to choose which to try
    • What clustering method will work best?

    View Slide

  11. How can we
    split our
    customers into
    different groups
    to market to?
    How can we run
    a clustering
    algorithm to
    segment
    customer data?
    Business
    question
    Data science
    question
    A k-means
    clustering found
    3 distinct
    groups
    Data science
    answer
    Business answer
    Here are 3 types
    of customers:
    new, high
    spending,
    commercial
    Domain knowledge
    - Renee Teate, @BecomingDataSci
    Skills:
    • Communication
    • Empathy
    • Understanding your data (where it lives, built-in assumptions, edge cases)

    View Slide

  12. How Do I Become a Data Scientist?

    View Slide

  13. “Fake data scientists”

    View Slide

  14. “Must know” lists

    View Slide

  15. You don’t need to know everything

    View Slide

  16. How to find a data science job
    Create a portfolio
    Expand your network
    Find the right jobs
    Tailor your application

    View Slide

  17. Creating a Portfolio

    View Slide

  18. What is a portfolio?
    A public body of work
    that illustrates your
    data science skills

    View Slide

  19. How?

    View Slide

  20. Dataset -> Question

    View Slide

  21. Dataset -> Question

    View Slide

  22. Question -> Dataset
    http://varianceexplained.org/r/trump-tweets/

    View Slide

  23. Tip 1: Include visualizations
    https://hackernoon.com/more-than-a-million-pro-repeal-net-neutrality-comments-were-likely-faked-e9f0e3ed36a6

    View Slide

  24. Tip 2: choose a topic you’re excited about
    https://masalmon.eu/2018/01/01/sortinghat/

    View Slide

  25. Tip 3: Limit your scope
    https://kkulma.github.io/2017-08-13-friendships-among-top-r-twitterers/

    View Slide

  26. Making progress
    Inspired by bit.ly/drob-rstudio-2019
    Less valuable More valuable
    Idea Getting data Cleaning Exploratory Final result
    Modeling
    Less valuable More valuable
    Work only on
    your computer
    Work online
    (GitHub, Blog, Kaggle)
    How I used to think about analyses
    How I think about analyses now

    View Slide

  27. The full process

    View Slide

  28. Put it on GitHub

    View Slide

  29. Where?
    ➔ Easy & quick to set up
    ➔ Organic traffic (medium)
    ➔ Less customizability/control

    View Slide

  30. Where?
    ➔ Complete control
    ➔ Always free
    ➔ Little longer to set-up
    ➔ May get stuck debugging issues

    View Slide

  31. Explain your analysis
    https://theambitiouseconomist.com/an-analysis-of-the-gender-wage-gap-in-australia/

    View Slide

  32. Teach a concept
    https://juliasilge.com/blog/stack-overflow-pca/

    View Slide

  33. Share your experience
    https://d4tagirl.com/2018/08/rstudio-conf-diversity-scholarships-for-the-win

    View Slide

  34. Give advice
    www.rladiesnyc.org/post/2019-nyr-conference-tips/ towardsdatascience.com/prioritizing-data-science-work-936b3765fd45

    View Slide

  35. Expanding your Network

    View Slide

  36. How people get data science jobs
    http://www.rctatman.com/files/Tatman_2018_DataSciencePortfolios_DC.pdf

    View Slide

  37. Meetups – search on meetup.com

    View Slide

  38. Twitter

    View Slide

  39. Ask for help (use hashtags)

    View Slide

  40. Live tweet talks

    View Slide

  41. Share your work

    View Slide

  42. Share other people’s work

    View Slide

  43. Reaching out
    • Mentions their work
    • Offers a topic
    • Suggests a specific time (that’s
    limited)

    View Slide

  44. Reaching out

    View Slide

  45. Finding the right job

    View Slide

  46. Let go of the “data scientist” title
    “Think about how attached you are to the data
    scientist title. If you decide to not concern
    yourself with what you’re called and to instead
    focus on the work that you’re doing, you’ll
    have a lot more flexibility to find jobs.”
    - Jesse Mostipak, Managing Director of Data Science at Teaching Trust, Chapter 5

    View Slide

  47. Figure out your specialty
    https://www.linkedin.com/pulse/one-data-science-job-doesnt-fit-all-elena-grewal/

    View Slide

  48. You don’t need to meet all of the “requirements”

    View Slide

  49. You don’t need to meet all of the “requirements”

    View Slide

  50. Consider the type of company matters
    Criteria Massive tech Retailer Startup Mid tech Government
    Contractor
    Bureaucracy
    Freedom
    Salary
    Job security
    Chances to
    learn
    Chapter 2, How data science works at different companies

    View Slide

  51. Tailoring your Application

    View Slide

  52. Writing a good resume
    • Include your GitHub and blog
    • Use clear, consistent formatting
    • Embrace whitespace
    • Limit to one page

    View Slide

  53. Quantify your impact
    • Ran A/B tests on email campaigns
    • Conducted 20 A/B tests on email campaigns,
    resulting in a 35% increase in click rate and
    5% increase in attributed sales

    View Slide

  54. Relate to data science
    1. Domain knowledge
    • Communication skills from teaching or consulting
    • Working in the domain (e.g. sales executive -> sales data scientist)
    2. Mathematics & Statistics
    • Classes
    • Research
    3. Programming & databases
    • Excel, Survey Monkey, Google Analytics, Tableau, SQL
    • Personal projects

    View Slide

  55. Writing a good cover letter
    • Try to find the hiring manager
    name
    • Tie together your experience
    • Not a repeat of your resume
    • Focus on what you offer the
    company
    • Tailor to the company

    View Slide

  56. Conclusion

    View Slide

  57. Take-away points
    • You don’t need to know everything
    • There’s no such thing as a “fake data scientist”
    • Let go of the “data scientist” title
    • Focus on:
    • Creating a portfolio
    • Expanding your network
    • Finding the right job
    • Tailoring your application

    View Slide

  58. Thank you!
    hookedondata.org
    @robinson_es
    datascicareer.com
    40% off w/ mtpdcds19

    View Slide