During October 2025, N.C. developed a vector-based semantic search feature for the BU-Spark/ml-bpl-rag repository, focusing on enhancing query relevance within a PostgreSQL environment. Leveraging Python scripting and database management skills, N.C. integrated a pre-trained machine learning model to enable semantic search capabilities, allowing users to retrieve more contextually relevant results. The work included a ready-to-run script, lightweight documentation, and example queries to facilitate adoption and integration. By delivering a practical, business-focused feature rather than addressing bug fixes, N.C. demonstrated depth in applying machine learning to real-world database challenges, providing tangible improvements to the platform’s search performance.