The four heights
The same task, four distances: today's deadline, the next reviewer, the stuck moment, the pattern.
Execute — do the immediate task
+Assemble the quarterly environmental health report: merge the incident logs, insert the new…
Execute — do the immediate task
+Assemble the quarterly environmental health report: merge the incident logs, insert the new exposure trend chart for June, ensure citations are consistent, convert to PDF/A for archival, and send the packet to the director and regulatory contact with a request for sign-off by Tuesday noon.
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Improve — make it easier to accept
+Before I circulate the report to the director, make it easy to review — pull the three most…
Improve — make it easier to accept
+Before I circulate the report to the director, make it easy to review — pull the three most important findings onto the first page, label any recommended actions with owner and deadline, and highlight any data gaps that would make a regulator ask for clarification.
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Decide — diagnose the stuck moment
+I sent the draft quarterly report and the director replied that the exposure trend chart looks…
Decide — diagnose the stuck moment
+Director flagged an unclear data gap.
I sent the draft quarterly report and the director replied that the exposure trend chart looks incomplete and might prompt regulator questions. I can’t recreate the missing rows in the dataset quickly and I fear the regulator will request the raw log. Should I delay distribution and fix the dataset, annotate the report pointing out the gap and propose a remedial audit, or send it with a cover note promising a corrected appendix next week? I don’t know how strict the regulator will be and I’m worried about credibility loss.
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Become — change the pattern
+Every quarter we get called out on minor data inconsistencies that force addenda and erode trust…
Become — change the pattern
+Reports repeatedly flagged for small data issues.
Every quarter we get called out on minor data inconsistencies that force addenda and erode trust with the regulator. Compiling reports takes weeks because I’m cleaning records instead of preventing errors. Should we introduce a pre-report validation checklist, train field staff on common logging mistakes, or automate row-level checks before compilation? Which single habit will most reduce these post-release fixes?
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Where the evidence lives
Who was seen doing this, and what people really ask.
Cell BiologistComputational BiologistData Quality AnalystField BiologistMarine BiologistPharma Formulation Scientistalso: Compile reports on the daily activities
Software tasks in the LLOS Work Atlas come from evidence, never a feature list: careers attested to do the work, real job descriptions, and the questions people actually ask (with their view counts). Facets — feature, workflow, troubleshoot, administer, deploy, scale — are open metadata: the work decides, not a taxonomy.
Copyright © LLOS.ai · 2026 — original pedagogy, voice, and design — all rights reserved.
The rest of the map
Same library, five ways in.