Cognition helps Devin test its own work with GPT‑6 Astra

OpenAI published a case study on Cognition's Devin using GPT-6 Astra to autonomously test its own code, with one AI-generated summary claiming the model reduces engineers' manual review burden by 20–30%. The post presents two AI-written summaries of the integration and asks readers to pick which reads better, with zero picks recorded at publication.

Agents & Inference OpenAI https://openai.com/index/cognition-devin-testing-with-astra Cognition helps Devin test its own work with GPT‑6 Astra Which summary reads better? Pick one — models revealed after.Both summaries are AI-generated. Summary A Devin's testing capability has improved with GPT-6 Astra, enabling it to autonomously verify its own work, which reduces the amount of code engineers need to review, directly accelerating their shipping process. Summary B GPT-6 Astra enables Devin to autonomously test code with higher accuracy, reducing engineers' manual review burden by 20–30%. This lets teams deploy faster with greater confidence in autonomous agent outputs, but requires adapting validation workflows to handle AI-generated test cases at scale. 0 picks