◈
Response Correction & Training Studio
Real feedback from
ai-prompt-evaluator
: review failed ORM assistant answers, correct expected behavior, and route fixes.
⬇ Export Dataset
+ New Correction
Total Feedback
☰
0
Pending Review
◷
0
Prompt Updates
↻
0
Backend/Data Issues
!
0
⌕
⚙ Filter
Feedback Queue
0 Pending
Sort: ID Asc
Sort: ID Desc
Sort: Impact
Sort: API Action
Showing 0 results
⟳ Refresh API Data
Selected Query Detail
Status:
Pending
ID:
-
Analysis:
-
👤 Original User Query
🤖 Observed Issue / Current Model Behavior
✅ User Expected Behavior / Correction JSON
ℹ Backend Feedback Metadata
Source
-
Output Type
JSON
Capability
-
Action
-
Impact
-
▧ Root Cause Classification
Issue Type / Capability
Affected Section
Capability Status
PARTIAL
NOT_APPLICABLE
UNKNOWN
Confidence
Low
Medium
High
Reason From Analysis API
▣ Training Decision
Needs Prompt Fix
Use proposed instruction to update parser/system prompt
Add to Training Dataset
Good example to improve model behavior
Needs Backend Fix
Parser is correct; investigate data retrieval/backend logic
Add Clarification Rule
User query is ambiguous / needs clarification flow
Ignore / Not Applicable
Not a prompt-training capability gap
☊ Review / Comments
Reviewer
Unassigned
Harshit
QA Reviewer
ML Engineer
Manager
Priority / Impact
LOW
MEDIUM
HIGH
⚡ Analyze Feedback Actions
0 Actions
Run Analysis
Submit Selected Actions
High Impact
0
Medium Impact
0
Prompt Updates
0
Ignored / Backend Issues
0
Action ID
Source Feedback
Capability
Action
Impact
Confidence
Reason
Proposed Instruction
Controls