A resume can say someone is fluent with AI. Their work habits tell you much more. This session gives leaders an observable definition of AI-native work, a way to find hidden operator-engineers, and a careful approach to evaluating real practice with consent and context.
The benchmark changes with the company. A portfolio operator, a finance leader, and a product builder should not be graded against the same model of good work.
How an AI-native operator actually works.
I walk through a prepared set of anonymized work artifacts, from the original request to the system that replaced several handoffs. The evidence shows how strong operators use agents to improve recurring work.
The session
Choose the role
I select work examples that match the roles represented in the audience.
Follow the evidence
A prepared case traces the work from prompt to process change and makes the observable behaviors clear.
Take the scorecard home
Leaders receive a private scorecard and a small, consent-based way to test it.
What people take home
- Define AI-native work through observable behavior instead of tool familiarity or self-reported fluency.
- Find the people already compressing handoffs, building small systems, and teaching the organization what works.
- Design a useful internal benchmark with consent, privacy, and human judgment built into the process.
The AI-Native Talent Scorecard
A private, adjustable scorecard for finding and developing internal builders. It starts a better human judgment process and produces no automated hiring verdict.
- Observable behaviors across AI-assisted and agentic work
- A role-specific benchmark worksheet
- A privacy and consent checklist for an internal pilot
Questions we work through
- Who inside the company is already AI-native?
- What evidence is more useful than a resume or a self-rating?
- How do we measure practice without creating surveillance theater?
A note on the talk
The strongest signal is often the handoff someone removed from the work.