GenAIOps Templates¶
The AI Factory provides GenAIOps templates for the GenAI-1 project type, enabling a full agentic AI lifecycle within a secure, private Azure environment.
GenAI-1 Baseline Architecture¶
Every GenAI-1 project deploys a secured baseline automatically:
| Service | Purpose |
|---|---|
| AI Foundry (Hub + Project) | Central AI development environment |
| AI Search | Vector search for RAG scenarios |
| AI Services / Azure OpenAI | LLM and embedding model hosting |
| Storage (×2) | Data lake + AI Foundry default storage |
| Key Vault | Secrets, keys, CMK support |
| Application Insights | Monitoring and observability |
| Private Networking | All services connected via private endpoints |
Optional services are added via feature flags — see Parameters.
Agentic Scenarios¶
The GenAI-1 template supports:
- RAG (Retrieval Augmented Generation): AI Search + AI Foundry + GPT-4o
- Agentic Workflows: AI Foundry Agents with network injection, Capacity Host (
enableAFoundryCaphost=true) - AI Gateway: APIM integration via
foundryApiManagementResourceId - Content Safety: Azure Content Safety service (
enableContentSafety=true)
Default Model Deployment¶
| Model | Default enabled | Default version |
|---|---|---|
gpt-4o |
✅ yes | 2024-11-20 |
text-embedding-3-large |
✅ yes (recommended for production RAG) | latest |
gpt-4o-mini |
❌ no | 2024-07-18 |
text-embedding-ada-002 |
❌ no | — |
text-embedding-3-small |
❌ no | — |
| Custom GPT-X | ❌ no | configurable |
All model deployments are configurable — see Parameters — Advanced Mode.
GenAI Project Team Personas¶
| Persona | Description |
|---|---|
p011_genai_team_lead |
Project lead, architect |
p012_genai_team_member_aifoundry |
AI Foundry specialist |
p013_genai_team_member_agentic |
Agentic workflow developer |
p014_genai_team_member_dataops |
Data pipeline engineer |
p015_genai_team_member_frontend |
Front-end / app developer |
Info
GenAIOps templates are located under copy_my_subfolders_to_my_grandparent/genaiops/.