Linear engineering leader Tom Moor argues that coding agents need context beyond the code, including user intent, previous decisions, related work and ownership. He frames product coherence as a challenge when implementation becomes faster.
Tom Moor describes an intake agent that groups requests, investigates issues and leaves customer information, related decisions and technical evidence for a subsequent coding agentAn agent handoff transfers an AI task, relevant context, authority, and responsibility between agents, people, sessions, or environments.. He emphasizes verificationAgent completion verification checks evidence that an agent achieved the requested outcome rather than merely producing intermediate work. and feeding continuous integration failures back to the agent.
Tom Moor describes a review process that brings earlier incidents, reverts, product decisions and agent sessions into context. He also discusses distilling lessons from completed changes into project instructionsAI agent memory is stored information that an agent can retrieve and use across steps, sessions, or changing contexts. while preventing those instructions from growing without limit.
Watch on YouTube
