Data Engineering: Building APIs, Data Pipelines, and Anomaly Detection Models to Power Financial Systems; Thirst for Technological Innovation: The Ethics of Water Consumption for Data Centers 64 views

My technical work and STS research paper are connected by a shared concern about the consequences of large-scale technological systems. The infrastructure that supports institutions can quietly externalize costs onto those with the least power to resist. My technical work showcased an internship experience in which I primarily worked on data engineering and machine learning to strengthen the financial systems that manage trillions of dollars in assets. My STS research examines how the same data centers consume enormous volumes of water and other resources, often in communities that had no say in the matter. Together, the two projects reveal that building efficient technical systems is only part of the engineer's responsibility, because the other part is evaluating where those systems sit in the world and who they affect. My technical work focuses on an internship experience that involved data center infrastructure. At XXXXXX, I worked on three main projects, each with a different discipline: data engineering, software engineering, and machine learning. My first project was to automate data processing pipelines to synchronize data containing millions of records. My other two projects were to develop Application Programming Interfaces (APIs) to process and route requests and to design a hybrid machine learning model to detect anomalous patterns in high-value wire transactions. Collectively, these systems contributed to a more scalable infrastructure for an institution managing approximately $13 trillion in assets by decreasing response times for retrieving information, generating data-processing insights, and increasing anomaly-detection accuracy compared to baseline approaches. My STS research paper explores a particular data center and its effects. xAI's Colossus supercomputer facility in South Memphis, Tennessee, was not an inevitable consequence of technological growth but a deliberately socially constructed outcome driven by unequal power dynamics. Using the Social Construction of Technology framework augmented by environmental justice principles, the paper traces how three groups, xAI, Memphis city officials, and local residents, assigned fundamentally different meanings to the facility. xAI's control over water consumption data and pre-arranged agreements with city utilities allowed a single corporate interpretation to dominate the public. The facility's location in a historically marginalized, low-income neighborhood and the systematic exclusion of residents from decision-making compounded the resulting environmental injustice. My paper explored these dynamics and discussed how information manipulation functions not as a regulatory gap but as a deliberate tool of closure, enabling corporate actors to exert control. Working on both projects sharpened my understanding of something that neither project could do on its own. At XXXXXX, I built pipelines and models for large-scale financial infrastructure. My STS research reminded me that the same infrastructure of this scale truly exists. It sits somewhere, consumes resources, and affects real people. The engineers who design these systems rarely encounter the communities that bear the consequences, and that distance can make it easy to forget they exist. Going forward, I intend to carry both perspectives into my work: the technical drive to build systems that are efficient and resilient, and the critical awareness to ask who bears the cost.

BS (Bachelor of Science)

data centers; AI; water consumption; environmental impacts

School of Engineering and Applied Science Bachelor of Science in Computer Science Technical Advisor: Briana Morrison STS Advisor: Benjamin Laugelli Technical Team Members: N/A

English

All rights reserved by the author (no additional license for public reuse)

2026-05-08