Cognition's New Benchmark Demands Real Production-Level Code Review Work

Cognition released FrontierCode on 8 June 2026, a coding benchmark where each task takes 40+ hours to build—far costlier than typical benchmarks. Rather than checking if code passes tests, it measures whether AI-written code meets actual production standards: clarity, architectural soundness, edge case handling. This reflects how human engineers truly evaluate code contributions.