{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/ddfcc1eb12874b11bb942b3488b7b1fb\" frameborder=\"0\" width=\"1280\" height=\"960\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":960,"width":1280,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":960,"thumbnail_width":1280,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/ddfcc1eb12874b11bb942b3488b7b1fb-a374b2f45a68910a.gif","duration":184.469,"title":"Automating Personalized Sales Outreach with Multi-Agent AI 🚀","description":"In this Loom, I explain why generic sales outreach fails and how our multi agent Sales Script Generator automates personalized, research backed outreach. The system runs an entire pipeline from research to final script, using Tavelli for company news, an ICP and Persona agent for target role and pain points, a copy strategy agent for tone, and an optimization agent to score and improve quality. Built in Python with Streamlit, it uses agentic research with multiple searches and Streamlit SessionState to pass structured data. I then ran a live demo targeting a senior thermal management specialist at Tesla for a high performance cooling system, and showed the optimized sealed script output. No direct action was requested from viewers."}