An Affordable Approach to Website Development with No-Code Content Management‬; Algorithmic Discrimination in Hiring: How U.S. Policies Fail to Prevent AI Bias

Profit with Purpose is a Contracted Independent Organization (CIO) at the University of Virginia that provides impact-driven consulting and investing to early stage companies prioritizing environmental, social, and governance (ESG) initiatives. The organization lacked an affordable platform to easily showcase and update their organization information on a semesterly basis, such as members, projects, partners, and opportunities to join. To solve this, I developed an affordable, maintainable, and user-friendly web application leveraging modern web technologies and cloud-based services. I implemented the application using React.js for the frontend and built the backend with two architectures: a cost-effective serverless architecture using Node.js and a scalable microservices architecture using Java Spring Boot and Docker. I utilized Firebase for a database, file storage, and authentication services. Firebase’s free tier allows up to 1 GiB database storage, 5 GB of file storage, and 50,000 monthly active users, which is more than enough given the frequency at which the website is updated. Finally, I deployed the site via Netlify. Netlify’s free tier supports up to 125,000 requests or 100 hours of serverless function runtime per month, which is more than enough for the website, given that updates generally only occur once per semester. The Profit with Purpose website is live and actively used to showcase projects, members, and communication information. It operates at a cost of only $10 per year, compared to Wix and Squarespace, which cost at least $204 and $192 per year, respectively. Future work may include performance optimization and additional testing to support long-term reliability.

Artificial intelligence (AI) is rapidly transforming modern recruitment practices, with over half of all companies in the United States relying on artificial intelligence to streamline the recruitment process. Trained algorithms introduce data-driven decision-making to help filter candidates based on a broader set of factors, such as skills, experience, behavioral patterns, and social media activity. The models are able to automate resume screening, act as virtual chatbots, conduct preliminary interviews, and evaluate cognitive abilities and behavioral patterns. While these systems are marketed as objective and efficient, I argue that current policies fail to prevent bias in hiring practices because they do not address the systemic biases embedded in algorithmic decision-making. Critical Race Theory (CRT) explains that historical discrimination is reinforced when past biases are rooted in algorithmic training data. Intersectionality further highlights how these tools compound disadvantages to disproportionately harm individuals at the intersection of multiple marginalized identities. The Social Construction of Technology (SCoT) framework explains that, despite these harms, AI persists because it benefits employers by cutting costs and increasing efficiency. To address the current gaps in policy, policymakers must implement mandatory AI bias audits, stronger federal oversight, rejection appeal processes, and enforceable transparency requirements. Without meaningful regulation, AI hiring tools continue to perpetuate historical inequalities and restrict equitable hiring processes.

The Profit with Purpose website and AI in Hiring do not inherently have many similarities. However, the projects are similar in a couple of ways. To start, both projects focus on disrupting the status quo. Many Contracted Independent Organizations at the University of Virginia use no-code platforms to design and deploy their websites, such as Squarespace and Wix. Even though these platforms are widely used, they have a key limitation in that they are not cost-effective. Squarespace and Wix cost at least $192 and $204 respectively. However, developing a custom-built website, while untraditional, may only cost as much as $10 per year. Similarly, the use of AI tools in hiring has become normalized across industries, yet these tools often perpetuate systemic bias by learning from and reinforcing patterns embedded in historical data. Both projects challenge widely accepted solutions because the standard approach often fails to meet the needs of those it’s supposed to serve.
The Profit with Purpose website and AI in Hiring also reinforce the principle that technology should be built to support the end user. With the Profit with Purpose website, I could have designed and deployed a Squarespace site as originally requested, but I instead listened to the organization’s needs and proposed a more affordable solution that they could maintain on their own. Similarly, in my STS research, I critique how AI hiring tools are often designed to optimize for efficiency or minimize cost for employers, with little regard for how the candidates are being evaluated. Many of these tools ignore how real people are affected. In both cases, I prioritize the human perspective and I design and evaluate systems based on the people who are affected most directly.

School of Engineering and Applied Science

Bachelor of Science in Computer Science

Technical Advisor: Briana Morrison

STS Advisor: Joshua Earle