New

Break down agent spend by model, person, and customer

Turn the agent traces you already collect into a clear breakdown of your AI spend, in your own warehouse.

Put this blueprint to work

Talk it through with the dltHub team and we will help you ship it to production.

Your coding agent

Claude Code, Cursor or Codex

operates through

the agentic layer

dltHub AI harness

Agentic primitives to build, run and fix pipelines

dltHub context catalog

Lineage, schema, data quality, governance, run state

grounds and runs

Usage pipelines

Claude Code
Codex
Cursor

the managed infra layer

Ingest usage, adapt as formats change, and join it to your billing data.

Dashboards

Ready-to-run dashboards that break spend down by model, person, and customer, in your warehouse.

Usage from Claude Code, Codex, and Cursor in. dltHub joins it to your bill. Spend by model, person, and customer out.

How it works

This Blueprint turns the agent traces you already collect into a clear breakdown of your AI spend, in your own warehouse.

Point it at your usage pipelines (Claude Code, Codex, and Cursor). dltHub ingests them, keeps working as the formats change, and joins the traces to the bill.

Ready-to-run dashboards then break spend down by model, by person, and by customer, so Finance can attribute it and Engineering can act on it. Because it runs on your warehouse, it can sit each customer's agent cost next to what that customer pays, so you see margin, not just cost.

Key features

Ingests usage from Claude Code, Codex, and Cursor, and adapts as formats change
Breaks spend down by model, person, and customer in ready-to-run dashboards
Joins agent cost to your billing data for true cost-to-serve and margin
Runs entirely in your own data warehouse, so no data leaves

How to get it

The standard agent-spend dashboards are included in dltHub.
Contact dltHub to get them.

Pricing

Standard: $1,190/month.
Custom: we build custom agent-observability dashboards together with our customers. Set up a 30-minute scoping call if you are interested.

Built by

dltHub

dltHub agent-spend dashboard, broken down by model, person, and customer

Put this blueprint to work

Talk it through with the dltHub team and we will help you ship it to production.

Find more blueprints

How it works

This Blueprint turns the agent traces you already collect into a clear breakdown of your AI spend, in your own warehouse.

Point it at your usage pipelines (Claude Code, Codex, and Cursor). dltHub ingests them, keeps working as the formats change, and joins the traces to the bill.