JevMade

Sign in
← Back to guides

JevMade field notes / Agency operations report

Sort agency work while keeping people in charge

Shahed Islam describes two agency trials with Jev, alongside eight proposed uses, showing how missing evidence and mismatched records limit automatic sorting.

Original by Shahed IslamAgent workflows
Leave a noteWhat would you try with this? : Sort agency work while keeping people in charge

Only you can see your notes.

Listen to this guide

JevMade’s plain-English explanation

0:00 /

Our summary

An agency repeatedly sorts meetings, messages and unfinished tasks. Shahed Islam describes using Jev to pick from real client and project lists, or judge whether task records show completion. The model does not write replies or finish the work; people and the surrounding software retain those responsibilities.

Islam reports testing meeting tags on 303 previously labeled meetings and reviewing 26 tasks. Restricting projects to the chosen client reduces possible mismatches. Task checks use comments and subtasks, and only recommend a status. Half the early task reviews lacked enough evidence to decide whether work was done.

Only two of the ten ideas were tested, and the results are the author's account, not independent measurements. Old human labels were disputed, client-project mismatches needed repairs, and a date rule reduced the review queue. High confidence cannot correct missing records or override a person's earlier decision.

Key takeaways

  1. Try suggestions on previously handled work without writing changes back; compare them with the actual records.
  2. Offer projects belonging to the selected client, then check that every stored client-project link still agrees.
  3. Treat missing completion evidence as a reason for review, not permission to close a task or judge an employee.

Islam reports about 97% client tagging on 303 meetings, with disputed old labels, and 29 correct project matches out of 33. His 3,924 meeting-tagging calls cost $0.41. The 85% and 60% decision rules are example policies, not empirical accuracy. Task checks are advisory and human choices are not overwritten. This vendor article also markets Agency Control Tower; eight other uses remain proposals.

ManagedCoder · Original published

Read the original guide Opens the author’s site in a new tab.

Your note

Only you can see this. Clear the text and save to remove it.

Keep what you save

Sign in to keep bookmarks and private notes on every device. We email you a link; there is no password, and the same link creates your account.