Build your AI apps and agents, hosted in France
A low-code engine powered by Yao, installed and maintained by DINAO. Model your data, generate your APIs, and orchestrate your AI agents — all on our French servers, never with a third-party vendor.
What is Yao?
Yao is an open-source low-code application engine distributed as a single binary, designed to build both web applications and AI agents. You describe its data models in JSON: Yao then automatically generates the REST APIs, CRUD, and interfaces, while business logic is attached via TypeScript hooks ("Create" and "Next").
On the AI side, Yao offers several execution modes: direct integration of LLMs (OpenAI, Anthropic, Mistral…), sandboxed CLI agents executed in isolated containers, or purely deterministic hooks. The engine features native support for the MCP protocol, vector search with embeddings, a multi-level memory system, and retrieval-augmented generation (RAG / GraphRAG).
Designed for production, Yao handles database modeling (SQLite or PostgreSQL), server-rendered component pages (SUI), OpenTelemetry observability, and simple Docker deployment, all without imposing a Node.js or Python stack to maintain.
Host Yao at DINAO
Resource tiers compatible with Yao prerequisites (minimum 1 vCPU / 512 Mo / 2 Go). Hosted in France, fully managed.
- 1 dedicated vCPU
- 2 Go RAM
- 20 GB NVMe
- Daily backups
- Managed & monitored by DINAO
- 2 dedicated vCPU
- 4 Go RAM
- 40 GB NVMe
- Daily backups
- Managed & monitored by DINAO
- 4 dedicated vCPU
- 8 Go RAM
- 80 GB NVMe
- Daily backups
- Managed & monitored by DINAO
- 8 dedicated vCPU
- 16 Go RAM
- 160 GB NVMe
- Daily backups
- Managed & monitored by DINAO
This application uses AI
The container hosts the application, not the AI engine (which requires dedicated GPUs) : inference is handled externally, with your own provider key. Prioritize a sovereign engine — Mistral AI, NumSpot, Scaleway, or OVHcloud AI Endpoints (France, GDPR) ; an international provider (OpenAI, Anthropic…) only if a specific capability requires it. Inference subscriptions are not included in hosting.
Technical details
You might be wondering…
Is Yao a real low-code tool?
Yes: you describe your data models in JSON and Yao automatically generates REST APIs and CRUD. For complex cases, you retain control with TypeScript hooks — no proprietary limits.
Can I connect my own AI models?
Yes. Yao works with an external LLM API (OpenAI, Anthropic, Mistral) or with an open-weight model hosted locally on your DINAO instance to remain fully sovereign.
Where is the data hosted?
On DINAO's infrastructure in France, in one of the available data centers. Your data and code never leave the territory.
Do I need a team of developers?
Not to get started: the low-code approach significantly accelerates application creation. A developer can still extend the logic via TypeScript hooks and the MCP protocol.
Can I change my plan or export my data?
Yes. You can upgrade or downgrade at any time, and your models and data remain exportable — no vendor lock-in.
