A six-month study of three emergency departments piloting LLM-assisted triage. We document where copilots reduced documentation load, where they introduced new failure modes, and the review protocols clinicians asked for.
Playbook · Science & Research
A six-pass sequential prompting framework for evaluating technical submissions
A structured multi-pass evaluation architecture in which each prompt targets a distinct analytical dimension — design rationale, prompt construction, and inter-pass consistency findings.