The four heights
The same task, four distances: today's deadline, the next reviewer, the stuck moment, the pattern.
Execute — do the immediate task
+Translate last quarter's pilot reactor yields into an industrial process estimate for the…
Execute — do the immediate task
+Translate last quarter's pilot reactor yields into an industrial process estimate for the production team. Take the pilot run data, convert per‑kg reagent consumption and energy use to a 1‑ton output using the provided scale factors, calculate expected yield loss at scale using a 3% derating factor, and produce a clear sheet showing per‑ton reagent needs and a weekly throughput plan for the plant operations meeting on Wednesday.
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 give this to production, make the conversion clear to non-research staff. Put the per-ton…
Improve — make it easier to accept
+Before I give this to production, make the conversion clear to non-research staff. Put the per-ton reagent and energy figures up top, explain where the 3% scale derate came from and how sensitive the total cost is to a 1% change, and flag the three assumptions that would make a production manager hesitate to commit to the schedule.
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 applied the scale factors to convert pilot data to a one-ton basis and the reagent totals are far…
Decide — diagnose the stuck moment
+My scale factor produced hugely different reagent totals than last run.
I applied the scale factors to convert pilot data to a one-ton basis and the reagent totals are far off compared with the last conversion I did. The scale multipliers are the same numbers I used before, but some per-run entries look duplicated. I'm worried I accidentally scaled an already-scaled column. What checks will quickly reveal whether I double-scaled data, and how do I fix it while preserving an audit trail so production trusts the numbers?
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Become — change the pattern
+We repeatedly waste time reconciling pilot-to-plant conversions because every engineer uses a…
Become — change the pattern
+Every scale-up we recreate conversions and lose trust with production.
We repeatedly waste time reconciling pilot-to-plant conversions because every engineer uses a different ad-hoc workbook and production questions the numbers. What workbook structure, naming conventions, and a minimal verification routine will stop repeated conversions and build credibility with operations?
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.
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.