Elizabeth Eardley, PhD, Developer in Berlin, Germany
Elizabeth is available for hire
Hire Elizabeth

Elizabeth Eardley, PhD

Verified Expert  in Engineering

Data Scientist and Machine Learning Developer

Berlin, Germany
Toptal Member Since
February 23, 2022

Elizabeth is a versatile data scientist, combining a strong scientific background from postdoctoral research with six years of industry experience, including Booking.com and Skyscanner. She excels at applying scientific techniques to data to support and automate decision-making, optimize products, and uncover actionable insights. Elizabeth has conducted extensive A/B testing and built and scaled multiple online experimentation platforms and optimization programs.


US-based SaaS Startup
Java, Product Management, Databricks, A/B Testing, Python, Product Roadmaps...
SQL, Python, Experimental Design, A/B Testing, Machine Learning, Statistics...
SQL, Python, Data Visualization, Experimental Design, Data Analysis...




Preferred Environment

Git, Jupyter, IntelliJ IDEA, Databricks, Snowflake, Python, MacOS

The most amazing...

...thing I developed is a patented statistical technique that uses causal inference methods to quickly and accurately detect software bugs and metric degradations.

Work Experience

Head of Data Science

2019 - PRESENT
US-based SaaS Startup
  • Led engineering and design teams to deliver projects from discovery and design through implementation and release. Projects included false discovery rate control, multiformat results and reporting exports, and automatic degradation detection.
  • Researched and designed new statistical algorithms and causal inference methods. Served as the lead inventor on the US patent for a statistical technique to accurately detect poor-performing software changes as quickly as possible.
  • Provided client consultations and support in experimental design. Led internal training sessions and initiatives to accelerate a culture of data-driven decision-making.
  • Delivered a wide variety of ad hoc data science support across all departments, including analyses, insights, and reporting.
Technologies: Java, Product Management, Databricks, A/B Testing, Python, Product Roadmaps, Customer Support, Public Speaking, Statistics, US Patent Process, Data Science, Hypothesis Testing, Causal Inference, Bayesian Inference & Modeling, Data Inference, Leadership, Data Analysis, Statistical Methods

Senior Data Scientist

2017 - 2019
  • Led the research and development of numerous extensions and improvements to the internal experimentation platform, including false discovery rate controls, guided power analyses, and in-depth reporting on experiment results.
  • Supported the design, implementation, and analyses of dozens of A/B tests across product, engineering, and marketing.
  • Ran internal training sessions and initiatives to grow the internal culture of data-driven decision-making and scale experimentation across the organization.
Technologies: SQL, Python, Experimental Design, A/B Testing, Machine Learning, Statistics, Redshift, Google Analytics, Mode Analytics, Mixpanel, Data Science, Data Analysis, Statistical Methods

Data Scientist

2016 - 2017
  • Developed productionized machine learning models, such as predicting the intent of a visitor to serve an optimal version of the product for the visitor's needs and preferences.
  • Designed, implemented, and analyzed A/B tests, driving numerous product decisions and influencing the long-term plans of multiple product engineering teams.
  • Measured the incremental value of loyalty programs and membership types using quasi-experimentation techniques like difference-in-differences, cohort analysis, regression discontinuity models, propensity score matching, and instrumental variables.
Technologies: SQL, Python, Data Visualization, Experimental Design, Data Analysis, Machine Learning, Reporting, Data Communication, Data Science, Apache Hive, Hadoop, A/B Testing

Postdoctoral Researcher

2015 - 2016
University of St Andrews
  • Applied modeling and machine learning techniques (classification and regression) to large numerical simulations of our universe to predict unobservable features of real galaxies.
  • Developed a novel method to extract valuable signals from the noisy and imperfect data of observed galaxy spectra.
  • Found statistically significant correlations between dark matter properties and their geometric environments. The results were published in peer-reviewed scientific journals.
Technologies: Machine Learning, Computational Physics, Python, Data Science, Statistical Methods, Linear Regression, Classification Algorithms

Analysis of the Effect of the Randomization Unit in Online Experiments

I identified a key driver of inaccuracies in a company's internal data analyses and analyzed and quantified its impact. I drove a meaningful shift in the internal processes and the company's approach to testing by communicating the research and findings, conducting training, and adjusting the infrastructure and tooling.

Published Work

I have published 13 articles as of January 2022. My academic research has centered around applying statistical techniques, including Gaussian processes, machine learning, and significance testing, to data from large-scale galaxy surveys to investigate and test scientific hypotheses.
2011 - 2015

PhD in Astrophysics

University of Edinburgh - Edinburgh, Scotland, UK

2007 - 2011

Master's Degree in Physics

University of Oxford - Oxford, England, UK


Slack, Git, Jupyter, IntelliJ IDEA, Jira, Google Analytics


Data Science


SQL, Python, Snowflake, R, Fortran, Java




Databricks, MacOS, Mixpanel


Redshift, Apache Hive


Statistics, Data Analysis, Data Inference, Hypothesis Testing, Experimental Design, A/B Testing, Mathematics, Physics, Mode Analytics, Segment, Computational Physics, Statistical Methods, Scientific Computing, Technical Writing, Research, Machine Learning, Data Visualization, Reporting, Product Management, Product Roadmaps, Customer Support, Public Speaking, US Patent Process, Causal Inference, Bayesian Inference & Modeling, Data Communication, Classification Algorithms, Linear Regression, Statistical Significance, Leadership

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.


Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring