flâneur

Jamie Davies on X: "Ask Attio: A technical look at our new agent" / X

x.com · 1,191 words · saved by 1 readers

To view keyboard shortcuts, press question mark View keyboard shortcuts Home Explore Notifications Chat Grok Premium Bookmarks Creator Studio Articles Profile More Post floguo @floguo Article See new posts Conversation Jamie Davies @viralpickaxe Ask Attio: A technical look at our new agent 6 7 124 14K Yesterday we shipped Ask Attio, the AI agent built into @attio. Ask Attio is designed to work directly with CRM data, understand the product context a user is operating in, and take action where appropriate. It’s not a standalone chat interface, but a system embedded into the core of the product. Building this required more than integrating an LLM. We spent time designing the underlying architecture that makes an assistant thrive in your CRM: non-linear conversations, dynamic capabilities, automatic context inference, and streaming output that can render structured, interactive UI. Here are some of the problems we encountered while building Ask Attio, and the systems we put in place to so

Yesterday we shipped Ask Attio, the AI agent built into @attio. Ask Attio is designed to work directly with CRM data, understand the product context a user is operating in, and take action where appropriate. It’s not a standalone chat interface, but a system embedded into the core of the product. Building this required more than integrating an LLM. We spent time designing the underlying architecture that makes an assistant thrive in your CRM: non-linear conversations, dynamic capabilities, automatic context inference, and streaming output that can render structured, interactive UI. Here…

related reading