Akka tested a spec-driven AI porting workflow across 65 open-source projects:
➡️ 9.41B tokens used over 99.3 hours
➡️ 57 ports delivered fewer lines of code or better performance.
See how the workflow works, the Sonnet vs. Opus speed/token trade-offs, and why infrastructure projects often took a performance hit.
📖 Read more: https://www.infoq.com/news/2026/10/ai-spec-driven-delivery/