Senior Machine Learning Scientist - ABU

Role Description: About Us

At Booking.com, data drives our decisions. Technology is at our core. And innovation is everywhere. But our company is more than datasets, lines of code or A/B tests. We’re the thrill of the first night in a new place. The excitement of the next morning. The friends you make. The journeys you take. The sights you see. And the food you sample. Through our products, partners and people, we make it easier for everyone to experience the world.

About the team

This opening is for the ABU ML team within Margin Management of the Accommodation Business Unit (ABU).

The ABU ML team develops causal machine learning systems that power one of Booking.com’s biggest customer acquisition channels. We predict which promotional investments will drive genuinely incremental demand and deploy these models in production at scale. The work involves uplift modeling, causal inference under marketplace interference, neural network design for structured data, and rigorous online experimentation.

The team actively contributes to the research community, our recent work “Converted Data is All You Need for Causal Optimization of e-Commerce Promotions” was published at ACM CIKM 2025. We encourage publishing and conference participation when the work advances the state of the art.

Role Description

As a Senior Machine Learning Scientist, you will design, build, and deploy uplift models and causal inference systems that allocate promotional spend across Booking.com’s accommodation marketplace. The role combines causal methodology, neural network architecture design, and production ML — with your work validated through large-scale A/B experiments. There are opportunities to publish applied research at top venues when the work contributes novel methodology.

Key Job Responsibilities and Duties:

• Design and deploy uplift models that estimate heterogeneous treatment effects, optimising incremental return on investment under budget constraints.
• Design and execute causal inference methodologies; including observational debiasing (IPW, doubly robust estimation), sensitivity analysis, and interference-aware evaluation to close the gap between offline metrics and online impact.
• Advance the team’s neural network architectures for uplift modeling on tabular data (attention mechanisms, multi-head designs, self-supervised pretraining), balancing model expressiveness with production latency requirements.
• Research marketplace interference and cannibalization; building frameworks to measure and correct for demand shifting when partial treatment is applied across competing properties.
• Develop offline evaluation methods that reliably predict online performance, accounting for biases introduced by non-stationary treatment policies and interference effects.
• Own models end-to-end; from research through A/B experimentation to production calibration.
• Collaborate cross-functionally with ML engineers on pipeline and serving design, with data scientists on feature engineering, and with product and business stakeholders on spend strategy and ROI trade-offs.
• Actively coach and mentor less experienced team members, setting technical direction and providing guidance on causal modeling best practices.

Qualifications & Skills:

• MSc or PhD (or equivalent experience) in a quantitative field such as Computer Science, Statistics, Economics, Econometrics, Operations Research, Mathematics, or Physics.
• Relevant professional or academic experience applying Machine Learning to business problems (typically MSc + 4 years, or PhD + 2 years).
• Advanced knowledge and experience in Causal Inference, Uplift Modeling, or Treatment Effect Estimation. Experience with heterogeneous treatment effects, interference / spillover effects, or policy learning is highly valued.
• Proven track record design