I design and direct AI systems for complex, evidence-based work.

What I bring

• Systems design: structure roles, inputs, outputs, and handoffs around an assignment.
• Evaluation: define repeatable drills, inspect evidence, and identify gaps.
• Human judgment: challenge assumptions and retain approval over consequential steps.

What I have built

• Mason: a catalog of 111 specialist role profiles and mission routing.
• Dojo: role curricula, skill passports, evidence review, and approved instruction updates.
• Arena: a view of recorded training state and sandbox traces.

How I approach an assignment

Frame the question. Select a focused team. Check the evidence. Challenge the assessment. Review the deliverable and retain a record of what changed.

Current scope

The built-in executor returns work contracts. External execution requires configured tools. The published example is illustrative; measured end-to-end performance and portable task forces remain future work.