SS

Platform OverviewRust Engine Active

Real-time context engineering telemetry, vector indexing, and pipeline performance metrics.

Qdrant HNSW: 18,450 vectors
Total Documents
142+12%
Ingested Chunks
18,450+8.4%
Collections
8Stable
Conversations
1,240+24%
Prompt Templates
24+2 new
Vector Embeddings
18,450+18.4k
Total Requests
342,910+32.1%
Avg Latency
142 ms-14ms

Requests & Latency Over Time

24-hour API request volume and average response time (ms)

Requests
Latency (ms)

LLM Provider Usage

Distribution of chat completions across model providers

OpenAI GPT-4o55%
Anthropic Claude 3.530%
Google Gemini 1.510%
Ollama Llama 3 (Local)5%

Retrieval Pipeline Latency (p95 ms)

Time spent in each sub-system during a hybrid context retrieval request

Total p95: 228 ms

System Status

100% Healthy

Gateway REST API

Gateway

4 ms

99.99%

PostgreSQL DB

Postgres

8 ms

99.95%

Redis Queue & Cache

Redis

2 ms

100.0%

Qdrant Vector Cluster

Qdrant

14 ms

99.98%

Background Worker Pool

Worker

11 ms

99.90%

Recent Activity Stream

Document Ingested

system_architecture_spec.md parsed & chunked into 84 vectors

5 mins ago

High-Volume Chat Session

Conversation conv_8f3a1d90 executed 8 messages via GPT-4o

12 mins ago

Eval Run Completed

RAG Retrieval Quality Suite v2.4 finished with 96.4% pass rate

45 mins ago

Ingestion Error

Failed to parse legacy_unsupported_binary.dat (Unsupported MIME type)

2 hours ago

Prompt Version Published

rag_contextual_qa updated to version v2.1.0

4 hours ago