Real-Time Telemetry • Langfuse • SLIs
Observability & LLMOps Console
Live system telemetry, multi-agent execution tracing, vector store cache analytics, and Langfuse LLMOps instrumentation.
Backend StatusOffline
Python Server Runtime
LLM FailoverGemini + Groq
Multi-Provider Redundancy
Search EngineTri-Search Grounding
Tavily / DDG / Google
Vector StoreChromaDB Persistent
Cosine Chunk Deduplication
Live Provider Health & Vector Storage Telemetry
Real-time connection status across LLM inference providers, search APIs, vector store cache, and Langfuse tracing instrumentation.
FastAPI Engine HealthDisconnected / Offline
Service:market-intelligence-agent
FastAPI Runtime:v0.162.0 (uvicorn)
API Ping Latency:Offline
LLM Orchestration ProvidersMulti-LLM Failover
Primary: Google Gemini (gemini-3.1-flash-lite)
Primary (Live)Failover: Groq (openai/gpt-oss-120b)
UnconfiguredAutomatic LangChain failover when Gemini encounters rate limits (429)
Grounding Search EngineTri-Search
Tri-Search: Tavily API + DuckDuckGo fallback + Google Grounding (Gemini) + Trafilatura Scraper
Tavily Search (Fallback to DDG)DuckDuckGo (Live)Google GroundingTrafilatura Scraper
Vector Store EnginePersistent Disk
ChromaDB Persistent Store on local server disk for semantic retrieval and debrief Q&A
Indexed Vector Chunks in Storage:Calculating...