D-04 · Engagement model
Scoped. Shipped. Handed over.
Small teams. Clean models. Efficient runtimes. Code you own. Each engagement is scoped, shipped and handed over — no retainer, no rent, no lock-in.
- Phase I
Domain discovery
We sit with the people doing the work. Map the workflow, the decisions, the evidence each step produces and the regulations it must satisfy. Identify where AI-first is safe, where Human-in-the-Loop is mandatory, and where AI should never go.
Output · Harness blueprint — workflows, tool surface, escalation policy, evaluation criteria, risk register.
- Phase II
Harness build
Implement in Rust. Build the tools, guardrails, observability, human-approval surfaces and evaluation suite. Model-agnostic via our GAISE abstraction — never locked to a single vendor.
Output · A working harness deployed to your environment, with a measurable baseline.
- Phase III
Supervised rollout
Run alongside your existing process. Humans review every output; the harness learns from every correction. Expand autonomy where the evidence supports it, contract it where it doesn't.
Output · A production-grade agentic workflow with auditable performance data and a defensible governance story.
- Phase IV
Ownership transfer
Train your team, hand over the code, stay on for as much or as little ongoing work as you want. The harness is yours. The IP is yours. The moat is yours.
Output · Full ownership — no retainer, no lock-in, no rent.
Questions
How does an engagement start?
With a free one-hour conversation with the team doing the work you want to augment, then a scoped discovery phase on one workflow.
Who owns the code?
You do. Each engagement ends with training your team and handing over the code: the harness, the IP and the moat are yours. No retainer, no rent, no lock-in.
How do you roll out safely?
The harness runs alongside your existing process with humans reviewing every output. Autonomy expands where the evidence supports it and contracts where it doesn't.