Reviewer View
System Trace
This view is meant for portfolio reviewers, not learners. It makes the demo legible as a working system by showing stored submissions, rubric-based scores, and the model and prompt trace behind each result.
What it proves
Learner submissions are persisted rather than mocked, so the demo can be reviewed as a real end-to-end flow.
Instructional signal
Scores are tied to the visible rubric, making the assessment logic easier to inspect and discuss with L&D or product stakeholders.
Product rigor
Prompt versions, model names, timestamps, and full results are exposed so the AI behavior is traceable rather than opaque.
Assessment records
Filtered to the single portfolio scenario used in this MVP. Each row shows the learner response, score, prompt version, and model used to generate the feedback.
| Timestamp | Learner response | Total score | Prompt version | Model | Action |
|---|---|---|---|---|---|
| Apr 7, 2026, 7:39 PM | Subject: Important Security Alert: Beware of Recent Phishing Emails Dear Team, We want to inform you that our IT department has identified multiple phishing emails being sent to… | 16/16 | evaluate_v1 | gpt-4.1-mini | Open trace |
| Apr 7, 2026, 4:41 PM | Subject: Urgent: Phishing Email Alert - Please Read Dear Northstar Staff, We have recently been alerted by our IT team to a series of phishing emails targeting our employees. Th… | 16/16 | evaluate_v1 | gpt-4.1-mini | Open trace |
| Apr 6, 2026, 5:15 PM | Subject: Watch out Hi everyone, There are some bad emails going around. Please be careful and do not trust everything you read. Let your manager know if something seems wrong.… | 6/16 | evaluate_v1 | gpt-4.1-mini | Open trace |
| Apr 6, 2026, 4:35 PM | Subject: Important: Be Alert for Suspicious Emails Dear Team, We have been alerted to several suspicious emails that appear to come from internal accounts. Some of these message… | 15/16 | evaluate_v1 | gpt-4.1-mini | Open trace |