Jack Thompson

Answered Best approach for prompt-injection defence in a user-facing app?

Assume injection succeeds and design so that it does not matter. Do not give the model a capability whose misuse you cannot tolerate — no unbounded tool calls, no raw SQL, no sending mail on its own authority. Everything else is defence in depth: mark untrusted content clearly, never let retrieved content look like instructions, validate outputs against a schema, and require a human confirm anything that writes. Delimiters alone stop the lazy attempt and nothing more.

Shared a bide Prepare for a Difficult Conversation Without Scripting It

Prepare for a hard talk with a report, peer or manager: separate facts from your story, open well, anticipate reactions and leave with a clear next step.

Shared a bide Triage profiler output and pick the one fix worth trying first

Paste cProfile, py-spy or Chrome profiler output and get the top hotspots by self time, an Amdahl-style ceiling, and one change to try first.