AI PLATFORM DESIGN
Video
by lizvladi
↗
Interactive architecture.
Request flow
Basic LLM
RAG
Agentic
Click any block to explore
Clients / Applications
Control Plane
Data Plane
Dependencies
Platform-wide
Request
Git
Argo CD
Helm
Controllers / operators
Gateway controller
Serving controller
Policy controller
GPU operator
Workload scheduler
AI Gateway
+
Model Inference
+
GPU Management
+
GPU
+
Identity Provider
+
Rate Limit Service
+
Guardrails
+
Inference Router
+
Vector DB
+
Object Storage
+
Model Registry
+
Observability
Security
Policy
Telemetry
AI Platform Design
Please enter password
Password
Enter
Password required