The single chatbot from v0.1 didn't scale to the kinds of questions our early users were asking. v0.2 introduced the multi-agent architecture that still underpins Curated today.
What shipped
- Google ADK (Agent Development Kit) backend — a new FastAPI service hosting purpose-built agents
- GIS Agent — geocoding, layer queries, spatial analysis, routing
- Salesforce Agent — pulls account and opportunity data and grounds answers in customer records
- Real-time streaming — answers arrive as they assemble, not all at once at the end
- WebSocket session management with heartbeat + exponential-backoff reconnection — production stability from day one
Why it mattered
A single LLM trying to do GIS, CRM, and conversational reasoning all at once gets each of them wrong. Specialists do better. v0.2 set up the orchestration so a question like "which Salesforce accounts are within 10 miles of our new Houston location?"* could route the spatial half to one agent and the CRM half to another, then compose the answer. The pattern is the same today — we just have more specialists.
For users
The transition was transparent. Conversations got faster and more accurate; users didn't have to know which specialist was doing the work.