Bias in AI Travel Narratives
↗An end-to-end study of cultural and representational bias in LLM-generated travel narratives, comparing human annotations with automated safety evaluations.
Projects.
Research, software, and systems engineering.
Showing all 6 projects.
Evolutionary Toxicity Control
↗An evolutionary search framework that iteratively refines LLM responses, combining toxicity evaluators with context-aware semantic scoring.
Tutor Scheduling Application
↗An RIT capstone with GCCIS automating tutor hour tracking, scheduling, and session management through backend APIs and an existing SQL database.
Fraudulent Job Detection
↗A classification pipeline for fraudulent job postings using text preprocessing, feature extraction, model tuning, and five-fold cross-validation.
Linux Performance Monitor
↗A lightweight application performance monitoring tool with interactive Python dashboards for real-time process metrics, log analysis, and anomaly detection.