AI Model Evaluation & Prompt Engineering — Personal Project
Hands-on evaluation and comparison of AI model outputs (Claude, Gemini, GPT) in real software development contexts. Activities included prompt design and testing for AI coding agents, Claude API integration for intelligent data categorization (FamilyBudget Pro project), comparative rating of model responses for accuracy, logic, and hallucination detection. Applied systematic Root Cause Analysis methodology to identify errors and inconsistencies in AI-generated code and text outputs.