Users say your assistant forgets what they told it, and sometimes remembers things wrong. How do you evaluate its long-term memory?

Single-session evals pass while users complain, because memory fails across sessions and at three separate stages. The answer is a planted-fact harness that names which stage lost the fact, plus the two tests teams skip: forgetting on request and leaking across users.

Updated Oct 2026 · Grounded in real Forward Deployed Engineer interview loops and written to a senior-engineer editorial bar.