Model at explainability pack
Amara Sorensen
Research & Development · Data Science & Modeling · RD5A Model Development
$20k
value
85%
confidence
19
people affected
1h 22m
age
2h 27m
sla
63%
progress
Currently held by
Validation Agent
An agent owns the next move
Raised by
Zara Sorensen
Requester of record
At station
Explainability pack
Workflow layer · station 5 of 8
Where this train is on the line
Evidence trail
Every station this unit has passed, who acted, and what they left behind
This is the readable trail. more
Cleared frame problem
-82mFeature Engineering Agent
Inputs, rule version and output hash written to the evidence log
Cleared engineer features
-66mFeature Engineering Agent
Spine control pack posted, three-way evidence attached
Cleared train
-49mFeature Engineering Agent
Spine control pack posted, three-way evidence attached
Cleared validate
-33mFeature Engineering Agent
Spine control pack posted, three-way evidence attached
Holding at explainability pack
-16mExplainability Agent
Inputs, rule version and output hash written to the evidence log
What happens next
The next station and its rule
Reviewer approve
run by Director, Data Science · Policy layer
Held for a named human when the unit exceeds the model decision envelope, when confidence falls below the floor, or when a policy clause is engaged.
approver Director, Data Science
Why the signal is what it is
Derived from time against commitment, not from the state
Green. This item is still inside the time it was committed to, and it is not close enough to that line for its materiality to pull it forward. Green says nothing about who holds it: an item can be green and sitting with a person.