<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/c89f2674ec8b4d68b8a7728072989567&quot; frameborder=&quot;0&quot; width=&quot;1658&quot; height=&quot;1243&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1243</height><width>1658</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1243</thumbnail_height><thumbnail_width>1658</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/c89f2674ec8b4d68b8a7728072989567-600de4f64a844218.gif</thumbnail_url><duration>170.549</duration><title>Smart Spidey AI for Instagram Fundraising Outreach</title><description>This Loom demonstrates the Smart Spidey AI system for connecting multiple Instagram fundraising campaigns and answering fundraisers’ questions without executives switching manually between campaigns. Ayush shows how to start a chat, select a campaign, and generate the first DM, with an option to enhance wording to 150, 200, or 250 words if needed. He explains the architecture: React with TypeScript on the frontend, Xspace on the backend, and Supabase plus VectorDB for embeddings, using campaign data split into chunks and normalized searching. He also covers end-to-end work including authentication, EC2 deployment, background jobs, controllers, schemas, and monitoring logs via GitHub actions.</description></oembed>