Sourcegraph

Teams at Stripe and Uber are scaling AI-generated code to thousands of PRs per week not by choosing better models, but by building 'Context First' architectures that give AI agents access beyond file-level context. The key insight is that confident but wrong AI code stems from insufficient context, and the companies pulling ahead are solving the context problem rather than the model problem.

The "Software Factory" is here, but why do AI agents still produce "confident, wrong code"?

The companies shipping 1,000 AI PRs a week aren't pulling ahead on model choice. They're pulling ahead due to context.

@ajaynz discovers how teams at Stripe and Uber are moving beyond file-level access to build "Context First" architectures. (1/2)

<p>The &quot;Software Factory&quot; is here, but why do AI agents still produce &quot;confident, wrong code&quot;?

The companies shipping 1,000 AI PRs a week aren't pulling ahead on model choice. They're pulling ahead due to context.

@ajaynz discovers how teams at Stripe and Uber are moving beyond file-level access to build &quot;Context First&quot; architectures. (1/2)</p>