Hey @pydantic team 👋
FastUI is one of the most genuinely interesting ideas in the Python ecosystem — using Pydantic models to define UI components declaratively, so your Python backend drives the frontend without writing JS. The philosophy is: define data shape, get interface.
I'm building Connectome, and I think the philosophical overlap is worth flagging.
Connectome's brain is Ora — a multi-agent system (10+ agents) that models who you are and generates personalized JSON screen specs in real-time. The React Native app renders whatever the agents produce. Think of it as: AI agents define the data shape of your current experience, and the frontend renders it.
It's a different execution from FastUI (agents rather than Pydantic models, React Native rather than web, personalization rather than general UI) but the core insight is the same: declarative, data-driven UI generation, just AI-driven rather than model-driven.
Stack: FastAPI + PostgreSQL + pgvector + Redis + Expo React Native
We're exploring whether agent-generated JSON specs could integrate with or learn from FastUI's component system. The FastUI component vocabulary is well-thought-out and applicable.
We're building as a DAO. Contributions earn CP via LTV-based evaluation. First 10 to reach 3,000 CP become Founding Stewards — permanent governance rights.
Wanted to surface this connection. If anyone on the Pydantic team is interested in where AI-driven declarative UI generation goes next — come see.
👾 https://t.me/ascensioncommunity
🌐 https://avielcarlos.github.io/ascension-site
— Avi / Aviel Carlos | @neoandromeda | carlosandromeda8@gmail.com
Hey @pydantic team 👋
FastUI is one of the most genuinely interesting ideas in the Python ecosystem — using Pydantic models to define UI components declaratively, so your Python backend drives the frontend without writing JS. The philosophy is: define data shape, get interface.
I'm building Connectome, and I think the philosophical overlap is worth flagging.
Connectome's brain is Ora — a multi-agent system (10+ agents) that models who you are and generates personalized JSON screen specs in real-time. The React Native app renders whatever the agents produce. Think of it as: AI agents define the data shape of your current experience, and the frontend renders it.
It's a different execution from FastUI (agents rather than Pydantic models, React Native rather than web, personalization rather than general UI) but the core insight is the same: declarative, data-driven UI generation, just AI-driven rather than model-driven.
Stack: FastAPI + PostgreSQL + pgvector + Redis + Expo React Native
We're exploring whether agent-generated JSON specs could integrate with or learn from FastUI's component system. The FastUI component vocabulary is well-thought-out and applicable.
We're building as a DAO. Contributions earn CP via LTV-based evaluation. First 10 to reach 3,000 CP become Founding Stewards — permanent governance rights.
Wanted to surface this connection. If anyone on the Pydantic team is interested in where AI-driven declarative UI generation goes next — come see.
👾 https://t.me/ascensioncommunity
🌐 https://avielcarlos.github.io/ascension-site
— Avi / Aviel Carlos | @neoandromeda | carlosandromeda8@gmail.com