Batch-first QA cockpit

Operator quality, evidence, and risk in one daily view.

This scaffold is wired to the current FastAPI contracts. It can render live scorecard and interaction details when backend IDs are configured, and it falls back to representative sample data while ingestion and evaluation pipelines are still being populated.

Overall score81.4
Interactions evaluated22
Critical issues1
Review queue1
Current evaluation pulse

Interaction snapshot

Critical issues present
Channel

call

Operator

Operator Name

Lead outcome

won

Latest confidence

88%

Scorecard

Weighted category performance

2026-07-26
Communication Quality84
Sales Effectiveness79
Responsiveness82
Accuracy and Compliance78
Process Discipline83
Backend contract

Endpoints this UI expects

  • GET /api/v1/operators/{operator_id}/scorecards/daily?date=2026-07-26
  • GET /api/v1/teams/summary/daily?date=2026-07-26
  • GET /api/v1/reviews/queue?date=2026-07-26
  • GET /api/v1/interactions/{thread_id}
  • POST /api/v1/interactions/{thread_id}/evaluate
  • POST /api/v1/evaluations/{evaluation_run_id}/review
  • POST /api/v1/import/calls
Team summary

Daily leaderboard and review pressure

Avg 76.5
Anastasia Smirnova

Score 84.0 across 11 interactions; review items 0.

Mikhail Petrov

Score 69.0 across 11 interactions; review items 1.

Review queue

Manager attention for the current batch

1 queued
Mikhail Petrov

chat on Bunker; confidence 54%.

late_follow_up, missed_customer_question

Risk themes

Top findings driving manager attention

missed customer question

3 interactions flagged in the current slice.

late follow up

2 interactions flagged in the current slice.

wrong availability

1 interactions flagged in the current slice.

Interaction details

What the review pane already knows

Thread ID

00000000-0000-0000-0000-000000000002

Evaluation run

00000000-0000-0000-0000-000000000004

Lead / room

Mummy

SLA status

Met or not applicable

Thread duration

252s

Operator link

00000000-0000-0000-0000-000000000001