{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/9a85df2fd6a147c08f70d6d38e88346c\" frameborder=\"0\" width=\"1920\" height=\"1440\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1440,"width":1920,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1440,"thumbnail_width":1920,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/9a85df2fd6a147c08f70d6d38e88346c-9ef1936d54adb5af.gif","duration":317.6,"title":"Context Engineering for AI Agent Experiences","description":"This Loom explains context engineering for AI agents, focusing on how to design an end to end scenario. It highlights three pillars: the AI instruction (role, teaching philosophy, thinking structure), the tools added (for example MCQ cards, Google Slides, images, webcam and screen snapshot, or external systems like CRM), and data and memory (carrying knowledge across sessions and fetching external data). The video uses a quantum mechanics tutoring example and contrasts it with a cold calling scenario, emphasizing how rubric based evaluation and end session settings fit into the configuration."}