Post

Agents don’t “learn” like humans. But they do respond to signals in the prompt history — especially in long sessions. That means your feedback helps shape what comes next.

When something works, say so: “Nice — this covers the main use case. Let’s just handle the null edge case too.” This encourages continuation, not correction. It keeps momentum high — and quality stable.

You can also course-correct gently: “Almost right. But this loop doesn’t handle timezones. Try again with that in mind.” The agent takes the correction and retains the confidence. That balance matters.