/ 02
telemetry
all tracks

 Track 02 · AI Telemetry & Governance

Know what your AI spend actually buys — per commit, vendor-neutral.

If the only honest answer to “what is the AI worth?” is a shrug, you don’t have a spend problem — you have a measurement gap. This track closes it with a vendor-neutral record that ties every model call to the commit it produced, so the value of your AI is a number you can defend, not a story you tell.


/ 02
the work

The pain

Finance sees a token bill, engineering sees activity, and nobody can draw a straight line between the two. When leadership asks what the AI is worth, the honest answer is a shrug.

And if you have platform-disclosure or EU-AI-Act exposure on the horizon, “we don’t really track that” is about to stop being acceptable. Relying on each vendor’s own dashboards leaves you with a fragmented record that nobody owns and no two providers report the same way.

What you get

A capture layer that intercepts every model call across every vendor into one analytics database, and dashboards that attribute AI token spend per commit and per release — drillable from a top-line number all the way down to a single prompt.

Vendor-agnostic by design, so the record doesn’t depend on any provider’s reporting. The same dataset that answers finance’s question is the one that satisfies a disclosure request — one source of truth, not a reconciliation exercise.

Powered by
Wood Fired Telemetry, the AI Telemetry Dashboards, and the Telemetry Query MCP. /tools
Shown, not claimed
You Can’t Ask an Agent to Log Itself, plus the live commit/source-line evidence on /practice.
Depth
From a one-week diagnostic audit, to a telemetry rollout, to a full on-prem governance stack built into your own infrastructure. scoped on the call

Other tracks: Trustworthy Agent Output · Agent Orchestration

/ 03
questions

Questions buyers ask first

Is the telemetry vendor-neutral, or locked to one model provider?

Vendor-neutral by design. The capture layer intercepts every model call across every vendor into one analytics database, so your record does not depend on any single provider’s reporting and no two dashboards disagree.

Do you train on our data or share it with other clients?

No. The telemetry runs against your AI usage for your engagement only. Your prompts, spend data, and analytics stay yours — nothing is used to train models or shared across clients.

Do you work remotely, or only around Baltimore?

Remote-friendly and worldwide. I am based in Baltimore, MD, but the rollout runs remotely, and the on-prem option is built into your own infrastructure.

How long is a typical engagement?

It ranges from a one-week diagnostic audit, to a telemetry rollout, to a full on-prem governance stack built into your infrastructure. The exact depth is scoped on a 30-minute call.

Does this help with EU AI Act or platform-disclosure exposure?

Yes. The same dataset that answers finance’s "what is the AI worth?" question is the one that satisfies a disclosure request — one source of truth that attributes spend per commit and per release, ready for disclosure, not a reconciliation exercise.

/ 04
next steps

Sounds like your week?

One 30-minute call. We pin down how deep this goes — a one-week diagnostic, a telemetry rollout, or an on-prem governance stack — and if budget needs shaping before a SoW, I’ll help you build the case. Engagements contract through Wood Fired Games.