AI Provider Telemetry & System Health
Inspection of provider abstraction layer (Gemini development / OpenAI production failover), token latency, and pgvector embeddings health.
Active Provider (Dev)LIVE
Google Gemini 1.5 Flash
Embedding Model: text-embedding-004 (768 dims)
Avg Latency: 395msSuccess Rate: 99.9%
Production TargetCONFIGURABLE
OpenAI GPT-4o-mini
Embedding Model: text-embedding-3-small (1536 dims)
Routing: AI_PROVIDER=openai
Infrastructure Components
PostgreSQL 16 Multi-Tenant EngineCONNECTED (Prisma ORM)
pgvector Vector Memory ExtensionENABLED (HNSW Index)
NestJS Modular Monolith APIPORT 3000 /api/docs