Search⌘J
AK

AI Agent Studio Overview

Keep tabs on agent accuracy, vector knowledge retrieval, and live conversation flow across production engines.

Total Active Agents
0+1 this wk
0 live in production
Vectorized Knowledge (RAG)
0 FAQs1024-dim
Neon pgvector HNSW indexing
Total Conversations Logged
0+18.4%
Across 0 sessions
Inference Speed & Health
240 ms+12% fast
Groq LPU + Cohere Command

Agent Query & Inference Flow

Live token calls and user interactions handled across Groq and Cohere engines.

Hover over any bar to view monthly query volume
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
1,420total queries

Total autonomous AI conversations captured and indexed across all active agents.

NEON PGVECTOR RAG HIT RATE
1,284RAG citations

90% of queries retrieved contextual embeddings.

1284 cached1420 total

Recent Autonomous Agent Activity

Live Stream Active
Customer Support AI08:45
Resolved user query on SaaS billing upgrade
180ms
Lead Qualifier Bot09:15
Captured enterprise lead email: tech@acme.corp
240ms
Docs & API Assistant10:05
Matched pgvector cosine similarity on HNSW index
210ms
BuckBuck General Agent10:40
Tested live playground widget in standalone mode
195ms
Monthly Query & Token Quota1.0M tokens
420k tokens
target: 1.0M tokens
42% of monthly compute allocation consumed.

Agent Fleet & Deployments

Configure persona prompts, test responses, and copy embed tags for any website.

Loading agents from Neon PostgreSQL...