“Kun Chen claims that using a fresh context window for an AI agent to review code is more effective at catching edge cases, as an agent reviewing in the same session it wrote the code is biased towards believing its work is correct.”