Senior Applied Researcher

Senior Applied Researcher will own well-defined marketplace efficiency problems end-to-end, from problem formalisation through modelling, deployment, and production monitoring. You will operate independently on complex problem areas, make explicit trade-offs between competing objectives, and partner closely with engineering to deliver robust, scalable solutions.

Conduct in-depth market research and analyze trends to identify opportunities for marketplace growth and optimization
Develop and refine metrics, dashboards, and KPIs to monitor market performance and guide strategic decisions
Generate and validate hypotheses to address key market challenges, presenting insights and actionable recommendations to stakeholders
Investigate anomalies and deviations in key metrics, identifying root causes and proposing solutions
Build predictive models and advanced analytics frameworks to forecast market trends and inform resource allocation

4–7 years of industry experience in data analytics, data science, applied research, or PhD + 2–4 years post-PhD in industry
End-to-end ownership of individual marketplace models, policies, or decision mechanisms
Strong judgment in trading off model fidelity, latency, robustness, and business impact
Proven ability to move from abstract problem formulation to production-grade decision logic
Hands-on experience collaborating with engineering teams on deployment, iteration, and monitoring
Demonstrated ability to reason about system behavior under uncertainty, delayed feedback, and partial observability
Modeling supply-demand dynamics as stochastic, partially observed, feedback-driven systems; formalizing unit economics, service quality, and LTV as state variables and objectives
Design and analysis of experiments under interference, spillovers, feedback loops, and adaptive policies (e.g., switchbacks, clustered interventions)
Practical experience with DiD, panel methods, synthetic control, double-robust estimators, and causal ML approaches in production settings
Ability to distinguish short-term experimental uplift from long-run equilibrium shifts and cumulative feedback effects using time-series and dynamic causal methods.