AI LR · Insights
Daily counts · per-template compare · live log tail
← Jobs
Transporters →
Designer
Days:
7
30
60
90
↻ Refresh
Daily counts
Day
Uploaded
DONE
FAILED
High-conf
Corrected (jobs)
Avg time (s)
Per-template comparison
(only DONE jobs)
Transporter
Template
v
Jobs
High
Med
Low
Corrected
Correction rate
Avg time (s)
To A/B compare a prompt revision, look at consecutive versions of the same (transporter, template):
fewer corrections + higher confidence
at the same job count = the new prompt is winning.
Log tail —
/var/log/multitenant/prod.out.log
Filter:
File:
out
err
Lines:
100
200
400
↻
auto-refresh 5s
loading…