Dmitriy Kolesnikov

Independent Verification as a Measurable Property of AI Coding Pipelines: Three Pre-Registered Studies

Evaluations of AI coding agents measure capability: can the agent solve the task? This paper studies an orthogonal property that decides whether an agent's output can enter a codebase without human inspection — whether the evidence accompanying a change is independent of the chan...

Independent Verification as a Measurable Property of AI Coding Pipelines: Three Pre-Registered Studies

Evaluations of AI coding agents measure capability: can the agent solve the task? This paper studies an orthogonal property that decides whether an agent's output can enter a codebase without human inspection — whether the evidence accompanying a change is independent of the chan...