{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/75f737fcf1984aa39259efa1ff594434\" frameborder=\"0\" width=\"1728\" height=\"1296\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1296,"width":1728,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1296,"thumbnail_width":1728,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/75f737fcf1984aa39259efa1ff594434-679229cec02e37ae.gif","duration":462.57,"title":"AI Powered Post-Incident RCA Draft Demo","description":"This Loom demonstrates an AI powered system that automatically generates structured post incident RCA draft reports from incident logs. Dean Anasuchitra explains the project structure, including app.py for workflow management, database.py for report generation and retrieval, AIService.py for Grok AI model communication and prompt creation, and LogPhaser.py and DeferPhaser.py for preprocessing log data. He then runs the app at a localhost URL, enters sample incident details such as title, affected system, timeline, uploads error paths and logs, and generates the RCA draft. Asma Humera highlights the history feature where every report is saved in an SQLite database for quick access, improving consistency and reducing manual effort."}