Tech Stack
AirflowNumpyPandasPostgresPythonPyTorchScikit-LearnSQLTensorflow
About the role
- Work cross-functionally to translate business problems into testable analyses and experiments.
- Dive deep into data to find key insights that would impact the business.
- Define and evolve trustworthy KPIs and guardrail metrics for user engagement, retention, and monetization.
- Develop and deploy machine learning models to make recommendations and influence the product roadmap.
- Define and build new ML features using text and multimodal embeddings and GenAI.
- Validate offline learnings with online outcomes through A/B testing; design, execute, and analyze experiments to prove product change attribution.
Requirements
- PhD degree; or M.S. +5 years in Computer Science, Mathematics, Electrical Engineering, Statistics, Economics or Operations Research.
- 3+ years professional experience applying statistical methods to real product/business problems at scale (leading analysis end-to-end and influencing decisions).
- Deep expertise in statistical inference and experimental design: hypothesis testing, power/sample size calculations, variance reduction, etc.
- Proficiency in causal inference methods.
- Proven ability to translate offline analysis into online impact.
- Fluency in the Python analytics stack (pandas, NumPy), statistical modeling (statsmodels or scikit-learn) and machine learning packages such as LightGBM and XGBoost.
- Strong experience with SQL (e.g. postgres, snowflake, etc).
- Preferred: 2+ years of professional experience with large-scale recommender systems (for news feeds, shopping, ads, music, etc).
- Preferred: Hands-on experience with orchestration/transformation tools (e.g. dbt and Airflow).
- Preferred: Experience with deep learning and familiarity with tools such as PyTorch or TensorFlow.
- Preferred: Hands-on development of products/tools incorporating GenAI, LLMs, RAG, and/or Agents.