Role Overview
We are looking for a Data Scientist to build predictive models for ad performance metrics such as click-through rate, conversion rate, cost per acquisition, and return on ad spend. You will develop simulation frameworks to forecast campaign outcomes and apply causal inference to separate true ad impact from correlation.
Responsibilities
- Build predictive models for ad performance metrics such as click-through rate, conversion rate, cost per acquisition, and return on ad spend.
- Develop simulation frameworks (Monte Carlo, scenario and what-if, budget allocation) to forecast campaign outcomes and support planning decisions before spend is committed.
- Apply causal inference and incrementality methods to separate true ad impact from correlation.
- Engineer features from large-scale impression, click, conversion, and spend data.
- Own the full model lifecycle: problem framing, data preparation, training, evaluation, deployment, monitoring, and retraining.
- Design and analyze experiments (A/B tests, holdouts) to validate model-driven decisions.
- Partner with engineering to serve models in production and with stakeholders to turn business questions into measurable targets.
Requirements
- Strong foundation in statistics and probability, and in traditional ML: regression, tree-based methods (XGBoost, LightGBM, CatBoost), clustering, and time-series forecasting.
- Expert-level Python (pandas, NumPy, scikit-learn, statsmodels) and strong SQL.
- Solid grounding in experimentation and causal inference: A/B testing, uplift and incrementality, confounding, and sample sizing.
- Hands-on experience with simulation or optimization methods.
- Experience taking models from notebook to production, including evaluation, monitoring, and retraining.
- Ability to quantify uncertainty and communicate assumptions clearly.
- Bachelor's degree required.
Skills
- Python
- SQL
- XGBoost
- LightGBM
- scikit-learn
Benefits
- Flexible schedule
- Health insurance
- Paid sick time
- Paid time off
- Provident Fund
- Work from home