MIHA GAZVODA
📍Amsterdam, Netherlands 📩 miha.gazvoda1@gmail.com 🏠 mihagazvoda.com 👤linkedin.com/in/mihagazvoda
E D U C A T I O N CONTROL SYSTEMS AND COMPUTER ENGINEERING (MASTER’S DEGREE) Faculty of Electrical Engineering | University of Ljubljana Oct 2015 – Feb 2017 • Recognition for Excellence • GPA 9.64 / 10
APPLIED PHYSICS Faculty of Mathematics and Physics | University of Ljubljana 2011 – 2015 S K I L L S • Python, R, SQL, git • (Bayesian) data analysis/modeling • A/B testing, Causal inference, ML, AI • Software development: packages, reactive dashboards • Data engineering TALKS / PUBLICATIONS • How to estimate and leverage correlation between metrics (plenary) | Booking.com Science & Analytics ‘25 • A reality check on NHST in A/B tests | Outperform magazine by Eppo (print) • A Bayesian Multilevel Model for Effective Proxy Metric Use in A/B tests | Conference on Digital Experimentation @ MIT ‘24 • The Science & Culture of Experimentation at Booking.com | DataScience@UL-FRI Meetup ‘24 • Rethink Significance, Boost Impact | Booking.com Science & Analytics ‘24 • Poster: Estimating impact of shipped experiments with Bayesian shrinkage | Booking.com Science & Analytics ‘24 • Thumbs up to Empirical Bayes | Booking.com Science & Analytics ‘23 • Tossing Around with Bayesian Inference | Python Pizza Hamburg ‘21 PYDATA LJUBLJANA & AMSTERDAM Co-organizer | Nov 2021 – Feb 2024 | W O R K E X P E R I E N C E DATA SCIENTIST Booking.com | Aug 2022 – Present • First data scientist in Attractions, now leading the craft. Mentor colleagues. • Above expectations performance evaluation in 2023 & 2024. • Attractions' experimentation ambassador: A/B testing strategy, quality, designs, and analyses. Developed frameworks for (1) estimating impact of shipped A/B tests with a hierarchical model, (2) proxy metric evaluation and effective use. • Led end-to-end development of data-related product features: (customer-centric) ranking, social proof models (product tags), sort by rating; significantly enhancing business performance measured with A/B tests. • Utilized causal inference in quasi-experimental designs to assess the performance of marketing campaigns or non-experimental product changes. • Co-developed a tool to design and analyze heterogeneous treatment effects in experiments at scale. DATA SCIENTIST Outfit7 | Apr 2019 – Jul 2022 • Development and maintenance of A/B testing infrastructure, visualizations, and statistics, internal R packages, and big data pipelines (Airflow). • Initiative and delivery of research-oriented data science projects: (Bayesian) A/B testing, causal inference, value of app features. • Mentoring less experienced colleagues. PART-TIME FREELANCE DATA SCIENTIST Harding Center for Risk Literacy | Sep 2019 – Dec 2019 • Synthesizing and cleaning datasets on antibiotics prescription. Development and evaluation of machine learning models to predict antibiotic needs in patients. DATA SCIENTIST Celtra | Oct 2018 – Mar 2019 • Transforming ambiguously defined business problems into tangible and actionable outputs. Developing metrics and key performance indicators. Producing efficiency, performance, and adoption analyses. • Supporting other teams with the data using Snowflake data warehouse and SQL. Creating reports and dashboards using Spark, Hive, and Shiny. • Infrastructure: Creation and maintenance of data pipelines. DATA SCIENTIST (STUDENT) Dia-Vit | Jan 2018 – Jun 2018 • Development of machine learning models, data visualization, statistical analysis and signal processing to calculate glucose level in blood using Python. DATA SCIENTIST (STUDENT) AI Laboratory, The Jožef Stefan Institute | Dec 2014 – Aug 2015 • Development of anomaly detection mathematical model of friction coefficient in oil drilling machine. |