video1.0<iframe src="https://www.loom.com/embed/d01124195f684860805798821234b0e6" frameborder="0" width="1728" height="1296" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>12961728Loomhttps://www.loom.com12961728https://cdn.loom.com/sessions/thumbnails/d01124195f684860805798821234b0e6-9deda38d0cbb7322.gif208.192Hi, I am Jeffrey Cabrera, and my project is Speech Therapy Image Codes. I transformed a streamlit number guessing game into an image based word game that uses an agentic workflow to select prompts, evaluate answers with Gemini 2.5, generate a follow up fill in the blank sentence, run a reliability check, and log results with guardrails. I demo exact match, near correct passes, and fully incorrect inputs using reliability and debug logs. I also built PyTest logic checks and an evaluation script with predefined cases. No specific action was requested from viewers.