Data Scientist

at JustWatch
Published December 26, 2022
Location Berlin, Germany
Category Data Science  
Job Type Full-time  


JustWatch offers data-driven movie marketing for the film industry. Based on user data from and our apps - now the largest streaming guide worldwide - we run data-driven digital marketing campaigns for Hollywood’s latest blockbusters, most popular shows as well as independent movies. We are also building a new proxy that measures movies and tv show success: we are seeking to create a new industry standard combining modern methods of measurement and machine learning.

Who we are looking for

We are looking for someone that shares our passion for movies and tv shows, with a result-driven mindset, that is able to effectively communicate complex topics to non-technical audiences, and with extensive knowledge in the field of predictive analytics.

We love challenging goals. We question everything. We are always learning, and we are excellent in execution. We believe in the 80/20 rule and try to apply common sense and pragmatism to everything.

Here's what you will do

  • Develop and iterate a model that uses extensive survey data to predict how many streamers will actually watch the movie or show in the end
  • Help develop an extensive benchmarking system for our surveys that spans multiple platforms and employ statistical methods to gain insights from survey results
  • Make sure we gather the right data points needed to achieve this and assist our survey team to improve their process
  • Develop reports with visualization, summarize findings, and communicate with the stakeholders

What you will need

  • Passion for movies and tv shows and eagerness to build a predictive model to forecast the performance of movies & tv shows on streaming services
  • Solid knowledge of statistics e.g. population vs sample, credible interval vs confidence interval, bootstrapping, hypothesis testing
  • Hands-on experience with data modeling/prediction techniques and tools:
  • Techniques: unsupervised and supervised Machine Learning, including clustering methods, predictive models, outlier detection, sentiment analysis, and recommender systems
  • Languages: Python
  • ML and Visualisation: sklearn, XGboost, pandas, AUTOML, Plotly, Altair, Seaborn
  • Model Explainability: SHAP, Explainerdashboard, EVALM
  • Interest in gathering insights from historical data and passion for predicting the future from this data in general
  • Ability to apply data cleaning, feature engineering, statistics, machine learning, and common sense to predict future behavior
  • Readiness and curiosity to learn new tools and methods
  • Team player mindset. You enjoy working with talented and inspiring people.
  • Ability to cope with imperfect data sets
  • Excellent English communication skills, with the ability to synthesize, simplify and explain complex topics to different types of audiences

If you feel like you are missing some of the hard skills – don’t worry, we are hiring for the attitude, values, and personality.

We embrace our entrepreneurial responsibilities. Through our 6-figure environmental and social impact budget, we offset every employee’s carbon footprint, give back to the community, support causes that matter, and have recently implemented a wellbeing and mental health programme.

Read more about our culture in our blog post and Talent page:

How to apply:

Show your interest in any way that works for you, we simplified the application process - send us whatever you have or just your Linkedin and a message explaining why you are a great fit for this role. We won’t look at one-click applications.