ARCHITECTURE Signal 436
Agentic fitness functions pair AI agents with versioned rubrics to evaluate judgment-heavy architectural concerns that deterministic rules cannot
↗
An InfoQ article proposes agentic fitness functions as a governed judgment layer that uses AI agents calibrated against ADRs, ownership metadata, and rubrics to assess evidence-bound but non-mechanically-decidable architectural risks such as boundary fidelity and semantic contract drift.
Deterministic fitness functions already guard measurable invariants like dependency direction and latency budgets, but they cannot judge whether a changed service interaction is intentional collaboration or accidental coupling. Agentic fitness functions fill that gap by returning structured verdicts with scores, confidence levels, and escalation guidance, giving teams continuous feedback on concerns that previously required periodic manual review. The approach explicitly keeps deterministic gates as the primary enforcement mechanism and escalates low-confidence or high-blast-radius outcomes to humans.
Written by elseif from the cluster below · every claim links back to a sourceThe three things worth knowing
Deterministic fitness functions remain the primary enforcement mechanism for measurable invariants such as dependency direction, contract shape, latency budgets, security posture, and policy checks.
Agentic fitness functions add value for evidence-bound but judgment-heavy concerns including boundary fidelity, semantic contract drift, workflow coupling, and stale ADR assumptions.
A production-ready implementation separates deterministic gates from agentic advisory signals, scopes evidence to the change, applies versioned rubrics, returns structured verdicts, and escalates low-confidence or high-blast-radius outcomes to humans.
THE CLUSTER