Sign in Subscribe

While we hurtle toward the unknown, make a coffee and spend some time in the zooo.

By The Zookeeper in OpenAI —

Devin Learns to Grade Its Own Homework

Cognition has updated its AI software engineer Devin to test its own code using GPT-6 Astra, with the stated ambition of reducing review fatigue for human developers and accelerating deployment cycles.

  • Cognition claims the integration will significantly cut down the time engineers spend reviewing generated code.
  • Leveraging GPT-6 Astra aims to close the loop between writing software and proving it actually works.
  • The ultimate goal is to let humans ship more code while doing substantially less actual engineering.

Why should I care? Could be big
Autonomous software testing that actually works remains rare enough to warrant cautious observation.

Read the original: Cognition helps Devin test its own work with GPT‑6 Astra

Subscribe to Ueno Zooo

The State of the Zoo, our weekly round-up of what actually mattered, is on its way. Join the list to get it first.
[email protected]
Subscribe