RAG Pipeline Data Labeling and Prompt Engineering (Uniapply Project)
TextText Generation
Developed a Retrieval-Augmented Generation (RAG) system for university eligibility advising by indexing and semantically structuring university admission data. Designed prompt engineering processes to extract and parse user-provided subject combinations and point scores using natural conversational input. Labeled and validated structured prompt responses for accuracy and grounded result generation.• Indexed and semantically labeled TCU admission data to support RAG workflows
• Developed and refined prompt templates for extracting structured text features from user input
• Validated AI-generated recommendations against expected eligibility and admission requirements
• Evaluated system outputs for answer faithfulness and effectiveness
Developed a Retrieval-Augmented Generation (RAG) system for university eligibility advising by indexing and semantically structuring university admission data. Designed prompt engineering processes to extract and parse user-provided subject combinations and point scores using natural conversational input. Labeled and validated structured prompt responses for accuracy and grounded result generation.• Indexed and semantically labeled TCU admission data to support RAG workflows • Developed and refined prompt templates for extracting structured text features from user input • Validated AI-generated recommendations against expected eligibility and admission requirements • Evaluated system outputs for answer faithfulness and effectiveness
2025 - Present