Back to all integrations

Automation

Lineman and Datadog

Ship spend, savings and token counts into Datadog as OTLP metrics, on your Datadog site's intake.

What this gets you

Lineman pushes your AI-coding spend, savings and token counts into Datadog as OTLP metrics, on your Datadog site's intake. Cost then sits on the same dashboards as the rest of your engineering telemetry, refreshed every 15 minutes. Company totals arrive alongside per-developer and per-repo spend breakdowns, so you can see which people and repositories the bill comes from. Datadog still lists its direct OTLP intake as Preview and recommends an Agent or Collector for production, so you can point Lineman at your own OpenTelemetry Collector and forward from there instead.

What people use it for

  • Cost on the dashboards you already watch

    Add lineman.spend and lineman.savings to the Datadog dashboards your team opens every day. AI-coding cost gets read next to the rest of your engineering telemetry instead of in a separate tool.

  • Attribute spend by developer and repo

    lineman.spend.by_developer is labelled with the developer's email, and lineman.spend.by_repo splits the same spend by repository. Amounts are in USD. Both are capped at the 100 highest-spending entries: Datadog caps a payload at 512 KiB, and an uncapped breakdown would eventually breach it and stop the company totals getting through as well.

  • Alert on spend, not just uptime

    The series are ordinary Datadog metrics, so a Datadog monitor can watch them the same way it watches any other. Token counts carry a token.type label of input, output or cache, so you can track cache tokens as well as the headline figure. Cache is reported at company level only.

More in Automation

Set up Datadog

The documentation walks through connecting it, what Lineman reads and writes, and the settings worth knowing about.

Read the setup guide

Looking for something else?

Every integration in one place, searchable by name and filterable by what it does.

View all integrations