Pipeline Traces

Granular latency and logic execution tracking.

Overview

Arivu operates via a complex LangGraph state machine. The Traces dashboard provides a visual map of exactly how much time each node took and the transformations applied to the data payload locally.

Node-Level Telemetry

Identify latency bottlenecks down to the exact pipeline boundary.

State Visualization

Monitor how the state dictionary morphs through generation and verification.

What You Can Inspect

Execution Loops

Visualize self-correction loops. When the query_verifier rejects SQL, the trace shows the round-trip back to sql_generator with the error context injected.

Token Usage

LLM token usage tracking per node. See exactly how many input/output tokens were consumed and calculate cost-per-query.

Latency Breakdown

A per-node millisecond breakdown: intake, generation, verification, execution, and response. Quickly spot if the LLM or the database is the bottleneck.

Pipeline trace data persists inside your local arivu.db (SQLite) or a Redis cluster if configured. Your telemetry never leaves your infrastructure.