OFFERINGS
What we build, and what we check.
One build practice for agent workflows on Google's Agent Development Kit. Two fixed-price diagnostics for the agents and AI spend you already have. And the open-source recorder the diagnostics run on.
SCOPED ON THE CALL · YOUR REPO · YOUR CLOUD
Agent workflows on Google ADK
Production agent workflows built on Google's Agent Development Kit. Context is fetched by bounded code, not chosen by the model; each step works with exactly what it was given and hands its output to the next.
- 01Bounded fetch
- 02Agent node
- 03Next layer
- 04Decision points
ALSO OFFERED
Diagnostics, and the instrument behind them.
Fixed price, fixed scope, delivered as files you own. Each one stands on its own, and each one is a sensible first step before a build.
DIAGNOSTIC · TWO WEEKS · $7,500
Agent Flight Check
A fixed-price audit of the agents, MCP servers and coding assistants in the agreed scope: inventory, permission map, ten days of tamper-evident records, ranked findings and a remediation plan.
See Agent Flight Check →DIAGNOSTIC · THREE WEEKS · $9,500
AI Spend Audit
Seat-level usage, AI-attributed commits and PRs by team, cost per active seat, and a renewal recommendation for your Copilot, Cursor and Claude spend.
See AI Spend Audit →OPEN SOURCE · APACHE-2.0 · FREE
Witness
The open-source MCP flight recorder the Flight Check runs on. Hash-chained records of every agent tool call, kept on your own machines.
See Witness →NOT SURE WHICH ONE
Tell us the problem instead.
Thirty minutes. We'll tell you which of these fits — including when the honest answer is none of them.
Book a call →