The four heights
The same task, four distances: today's deadline, the next reviewer, the stuck moment, the pattern.
Execute — do the immediate task
+We have to adjust the household survey for field teams to make data collection feasible. Send the…
Execute — do the immediate task
+We have to adjust the household survey for field teams to make data collection feasible. Send the revised instrument and the enumerator notes to Luis, the field supervisor, and to Mei, the PI, asking them to confirm which questions can be dropped and which must remain by Thursday so training can proceed.
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 finalize this practical version of the survey, make it easy for field reviewers: put…
Improve — make it easier to accept
+Before I finalize this practical version of the survey, make it easy for field reviewers: put estimated interview time and respondent burden at the top, move sensitive and complex modules into optional sections, mark questions that require translation or show cards, and flag anything that would make an enumerator hesitate to ask in the field.
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
+Enumerators just reported that the full questionnaire takes 35 minutes in practice, blowing our…
Decide — diagnose the stuck moment
+field teams reported interviews take 35 minutes, not 20
Enumerators just reported that the full questionnaire takes 35 minutes in practice, blowing our target of 20 and risking respondent fatigue. I cannot tell which modules are essential versus negotiable for analysis. What is the likeliest pruning that preserves primary outcomes and the best immediate cut to meet the 20-minute target?
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Become — change the pattern
+Across studies our field surveys keep exceeding target interview lengths, forcing late design cuts…
Become — change the pattern
+surveys repeatedly overrun planned interview time
Across studies our field surveys keep exceeding target interview lengths, forcing late design cuts and compromising measures. Is the root cause question creep, poor pretesting, or overambitious measurement? Recommend one sustainable change to keep instruments within fieldable time limits without losing core data quality.
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.