Amazon Advertising Data: An Automated AWS Pipeline; How Memory in History and Technology Shapes Perspectives on Safety and Security 136 views

The technical aspect of my thesis focuses on a advertising pipeline I created during an internship at Amazon. This involved the usage of various Amazon Web Services (AWS) tools to create a system that would take in advertising metrics as inputs, aggregate these based on predetermined metrics, and publish a business report that a finance team is able to analyze. In the technical report, I detail the high-level design of the pipeline, which involved two major sections, firstly the components of the pipeline, which is the infrastructure of AWS that the pipeline would be orchestrated upon, as well as the scheduling of the pipeline, which would determine the amount and frequency of resources needed to be dedicated to the pipeline to ensure its successful operation. I then analyze the low-level aspects of the pipeline, as well as the challenges presented in each part of this project-planning phase. The low-level explanation describes the data specifications the pipeline operates upon, which provided a motivation for the STS portion of the thesis. I then summarize the outcome of the successful implementation of the pipeline within the advertising team, and describe the work that will take place in the future to support the pipeline within the team. The STS portion of the thesis focuses on how the long-term retention of information as opposed to the either conscious or unconscious collective forgetfulness of it can have negative consequences for a society.

BS (Bachelor of Science)

data pipeline; advertising; amazon web services; AWS; memory; data retention; collective forgetfulness

English

2024-05-06