Agents forget across steps and sessions
An agent's working memory is its context window, which is finite and resets between runs. On long-horizon tasks it forgets earlier steps, repeats work, and loses the user's intent — so "agent…
Obstacles to building & operating agents, mapped to solutions
Agent engineering: obstacles & solutions
A living map of the obstacles to building and operating AI agents — memory, reliability, tool use, cost, and more — each linked to the solutions the field uses, with source articles.