Learn it

Performance Coaching Feedback

Rubric-anchored feedback on every real attempt, not just the ones a manager sees.

Ingests a real performance artefact — a call recording, a written pitch, a code submission — evaluates it against a criterion rubric, and returns specific feedback the learner can act on next attempt. Use this when the skill only improves through iterated feedback on real work, not practice exercises.

Shape

FEEDBACK ON REAL WORKreal artefact(call · brief · code)Scorerubricfeedbacklearnernext attemptimprovement over attempts

Operational dimensions

Human supervisor

Person oversees and intervenes by exception.

Event-triggered

Fires when an upstream condition occurs.

Low data gravity

Light state; replaceable any time.

Read-only inbound

Consumes external data; does not write back.

Inputs

  • real performance artefact (recording, text, submission)
  • criterion rubric with scoring dimensions
  • optional prior feedback history for the learner

Outputs

  • criterion-anchored feedback per rubric dimension
  • overall performance score
  • improvement signals aggregated across attempts over time

Mechanism

Ingests a real-world performance artefact, scores it against a criterion rubric, and returns structured feedback that the learner applies on the next attempt.

Why this is a primitive

Cannot be decomposed — the ingest-artefact → criterion-score → structured-feedback loop on real performance is one operation. It is not live instruction (the performance has already happened), not synthetic practice (the artefact is real), not mastery certification (the loop is ongoing).

Where it shows up

Revenue enablement — every sales call is transcribed and scored against a discovery rubric; reps receive feedback within minutes without manager listening to every call
Law school — student briefs scored against an argumentation rubric with specific line-level notes, not just a grade
Trades apprenticeship — trainee submits a photo of completed work; system scores against quality criteria and flags what to fix on the next job

Related primitives

Tags

AIlearningfeedbackhuman-collaborativestructured-data

See where it fits.

Primitives are configured into named solution shapes for each client’s domain. The fastest next step is a conversation about which shape fits your problem.

Start a conversation