Developer Builds AI Resume Screener That Strips Bias Before Human Review
A software developer has built an AI-powered blind resume screening tool designed to remove unconscious hiring bias by redacting personally identifiable information before any evaluation takes place. The system uses CrewAI, OpenAI's GPT-4o, and AWS to automatically strip names, email addresses, and other demographic signals from resumes before a two-agent pipeline assesses candidates. The developer originally planned to use AWS Bedrock but switched to OpenAI's gpt-4o-mini after encountering regional availability restrictions and latency problems. AWS IAM policies were rebuilt from scratch following the Principle of Least Privilege to ensure the system accessed only the specific data it needed, a step the developer describes as non-negotiable when handling candidate data. The fully event-driven pipeline delivers a structured evaluation report via email with no manual intervention required.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in