How we built Linear Agent

linear.app · 1,668 words · saved by 2 readers

The product and engineering decisions we made so Linear Agent can handle complex tasks without becoming unpredictable.

Most software is designed to behave consistently. Good engineering has typically meant shrinking the space of possible outcomes until the same action reliably produces the same result. AI inverts that logic, at least somewhat. Much of Linear Agent’s value lies in doing work we didn’t anticipate in ways we never explicitly defined; script its behavior too tightly, and we’d dilute the flexibility that makes it useful. So rather than engineering a fixed path for Linear Agent, we defined the boundaries within which it could find its own. We drew those boundaries in the agent’s system prompt,…

How we built Linear Agent

saved by

related reading