Data Scientist

Who you are

• You love to work with data: explore it, model it, improve its quality
• Deep grounding in statistics: you know which method fits which problem and can defend your assumptions, not just run the library defaults
• Fluent in Python (pandas, scikit-learn, NumPy) and SQL, with a track record of applying them to real business problems rather than toy datasets
• Hands-on experience taking ML and modern AI techniques from idea to a working solution that someone actually uses
• Practical experience with LLMs and RAG systems in production or near-production settings, including prompting, retrieval quality, and output evaluation
• Solid command of A/B testing: sample sizing, significance, common pitfalls, and knowing when an experiment is the wrong tool
• Working knowledge of performance marketing concepts such as CAC, ROAS, and attribution logic
• Project experience in at least one of: anomaly detection, trend analysis, marketing mix modeling, or multi-touch attribution
• Background in e-commerce, marketplaces, or other platform-based businesses, ideally with exposure to supply and demand dynamics
• Bonus: degree in mathematics, statistics, physics, computer science, or a related quantitative field

What the job involves

• We're looking for a Data Scientist who treats AI as a working tool, not a buzzword.
• You'll sit at the intersection of statistics, machine learning, and product: building predictive models, improving our LLM- and RAG-based systems, and running experiments that directly shape how our platform matches supply and demand.
• Your work won't end at a slide deck. You'll define the metrics, ship the analysis, and follow through until the impact shows up in the numbers
• Build, validate, and ship statistical and predictive models that directly inform pricing, matching, and growth decisions
• Develop and improve LLM-powered features, from retrieval-augmented generation (RAG) pipelines to applications of new AI technologies that open up product innovation
• Own the reliability of our AI features: design prompt and evaluation workflows, measure output quality, and catch regressions before users do
• Turn open questions into testable hypotheses and design experiments (e.g., A/B tests) that give clear, decision-ready answers
• Dig into funnels and user journeys to find drop-offs and friction points, and quantify where supply and demand can be better matched
• Team up with performance marketing to sharpen targeting, attribution, and campaign efficiency with data
• Define the KPIs that matter, build the dashboards and monitoring behind them (AWS QuickSight), and make business impact visible and measurable
• Keep your work transparent and traceable: document, prioritize, and communicate progress in Jira across product, engineering, and marketing
• Present findings to stakeholders as concrete recommendations, then stay involved until they're implemented