Amazon Applied Scientist II Advertising Incrementality Measurement 

United States, Colorado, Boulder 

108519967

Key job responsibilities
* Partner with economists and senior team members to drive science improvements and implement technical solutions at the state-of-the-art of machine learning and econometrics
* Partner with engineering and other science collaborators to design, implement, prototype, deploy, and maintain large-scale causal ML models.
* Carry out in-depth research and analysis exploring advertising-related data sets, including large sets of real-world experimental data, to understand advertiser behavior, highlight model improvement opportunities, and understand shortcomings and limitations.
* Define data quality standards for understanding typical behavior, capturing outliers, and detecting model performance issues.
* Work with product stakeholders to help improve our ability to provide quality measurement of advertising effectiveness for our customers.

- 3+ years of building models for business application experience
- PhD, or Master's degree and 4+ years of CS, CE, ML or related field experience
- Experience programming in Java, C++, Python or related language
- Experience with ML modeling toolkits such as sklearn, PyTorch, Tensorflow, etc.
- Experience with data exploration and querying technologies such as pandas, SQL, and Spark.
- Very strong written communication skills. An ideal candidate needs to be able to distill complex technical details into tangible insights for leaders and stakeholders.

- Experience with causal inference, A/B testing, experimentation, etc.
- Experience with MLOps practices for developing models in a production setting.
- Experience with causal ML modeling (e.g. double machine learning, uplift modeling)
- Familiarity with AWS systems (e.g. S3, EMR, Sagemaker).
- Experience with statistical visualization (e.g. ggplot2, plotly, Seaborn).
- Experience with digital advertising is a bonus, but not required.