Akka tests spec-driven AI porting across 65 open-source projects
Akka published an experiment in spec-driven AI-assisted software porting: it analyzed 65 open-source projects, generated specifications and implemented up to 10% of each project's functionality, with the first round taking 99.3 hours and 9.41 billion tokens, then selected 10 projects for full implementation based on system traits and measurable results. Akka says 57 of the 65 ports improved in lines of code or performance — Dify reportedly improved 143,333x (though the compared workloads differed) while Netflix Metaflow ran about 100x slower. Claude with Akka Specify handled implementation, testing and review; Sonnet averaged 61 minutes per port versus 120 for Opus, which used roughly 40% fewer tokens, and raising effort settings increased consumption without consistent efficiency gains.
It offers a public data point on how specifications and context affect the real delivery cost of AI coding agents.
Akka 利用 65 个开源项目，对一种基于规格说明的 AI 辅助软件移植工作流进行了测试"，评估内容包括：规格说明的结构、上下文、模型和工作量选择、自动化验证、token 消耗以及运行时性能。首轮测试耗时 99.3 小时，消耗了 94.1 亿个令牌。根据 Akka 的报告，在 65 个移植项目中，有 57 个在代码行数或性能方面实现了提升。
该实验分成了两个阶段。Akka" 对全部 65 个项目进行了分析，生成规格说明并实现了每个项目最多 10% 的功能。随后，他们根据系统特性和可量化的结果，从中选出 10 个项目进行了完整实现。交付管控工作流循环经历了需求发现、规格定义、移植、基准测试和优化等阶段。需求发现阶段对代码、模型、模式和运行时行为进行了分析，而 Claude 配合 Akka Specify" 则负责实现、测试和审查工作，然后通过一个通用的基准测试运行器，对测试、代码规模和延迟进行了比较。
#Akka#Akka Specify#Claude#Anthropic#GitHub Spec Kit#Dify#Netflix Metaflow