A Day in the Life of a Remote Data Scientist

Yanir Seroussi

yanirseroussi.com | @yanirseroussi | linkedin.com/in/yanirseroussi

Motivation

What people have got to realise is that houses are much cheaper in Tamworth, houses are much cheaper in Armidale, houses are much cheaper in Toowoomba.
- Barnaby Joyce urges aspiring home owners to look beyond Sydney following dire housing survey (2017)

Outline

  • Preliminaries
  • A day in the life
  • Pros and cons

Preliminaries

Defining data science (again)

...we can define data science as a field that deals with description, prediction, and causal inference from data in a manner that is both domain-independent and domain-aware, with the ultimate goal of supporting decisions.
- Me, Defining Data Science in 2018

My path so far

  • Haifa, Israel: Technion BSc, Intel, Qualcomm
  • Melbourne/Sydney: Monash PhD, Google, Giveable
  • Sydney: Next Commerce, consulting, own projects, Car Next Door
  • Ballina: Automattic

Ballina?


Automattic?

  • Main products: WordPress.com, Jetpack, WooCommerce
  • Fully distributed: 800+ employees, no central office
  • One of the few companies that hire remote data scientists
  • Longest held position since 2012

Further reading: My 10-step path to becoming a remote data scientist with Automattic

A day in the life

Morning boxland

Epic break time

Afternoon quiet work

Actual work examples

  • Recent: pipe and RR++
  • Previously: Elfbot and CEMAP
  • Others: data requests and products

Further reading: data.blog

True flextime: YMMV

  • My chronotype: morning person
  • Not a team lead = fewer meetings
  • Easy to work too little or too much

The joy of meetups


Pros and cons

For employees

Pros:

  • True flextime
  • Quiet time for deep asynchronous work
  • Free meetup trips
  • Learn from people with diverse backgrounds

Cons:

  • Hard to maintain work/life boundaries
  • Potential isolation
  • Long flights and jetlag
  • More cultural differences and sensitivities

Further reading: Reflections on remote data science work

For employers

Pros (if done well):

  • Happier employees
  • Calmer work
  • Hire anywhere
  • Spend less on offices and on-site perks

Cons (if done badly):

  • Isolated employees
  • Communication breakdowns
  • Less sense of control
  • Spend more on meetups and home-based perks

Conclusion

The future is distributed

You don't have to...

...work for a bank

...pay millions to live in "Sydney"

...move to Tamworth

The future is up to us

Ask recruiters and employers for remote options

Let your teams work remotely, at least part-time

Strive for asynchronous calm conditions

Questions?