New from the Lab·The Compass — an open moral reasoning standard for AI, tested across frontier modelsExplore →
Production AI Institute · PSF v1.1 open standard
AI Right-To-KnowAI Data Use IndexCheck My AI ToolsPolicy Change WatchAgent ReadinessPublic BenchmarkContact
Official Document · PAI-CPAP-001

CPAP Assessment Rubric

Certified Production AI Practitioner — transparent scoring criteria for portfolio assessment. Every dimension, every level, every criterion published openly.

Scoring Summary

D1
25
points
D2
25
points
D3
20
points
D4
15
points
D5
15
points
Total possible score100 points
Pass threshold70 / 100

Scoring Dimensions

D1

Production Deployment Evidence

25 pts

Demonstrated experience deploying AI systems to real production environments serving real users.

Exemplary23–25 pts

≥2 production deployments with documented scale (users/requests/revenue impact). Includes monitoring dashboards, incident history, and iterative improvement evidence. Clear before/after metrics.

Proficient18–22 pts

1 clear production deployment with some quantitative evidence. Monitoring exists. Some post-launch iteration documented. May lack scale metrics or long operational history.

Developing10–17 pts

Internal tool or limited beta deployment. Evidence is partial — screenshots, anecdote, or team-only usage. Limited operational history. Monitoring described but not evidenced.

Insufficient0–9 pts

No production deployment. Prototype, demo, or academic project only. No user impact evidence. Claims of production usage are unverifiable.

D2

PSF Application

25 pts

Correct application of the PAI Production Safety Framework across all 8 domains in a real deployment context.

Exemplary23–25 pts

All 8 PSF domains addressed with specific, concrete implementation evidence. Tradeoffs between controls are discussed. Evidence of framework evolution as the system matured.

Proficient18–22 pts

6–7 PSF domains addressed with clear implementation evidence. One or two domains may be lighter on detail. Shows understanding of why each control was chosen.

Developing10–17 pts

4–5 PSF domains addressed. Evidence is surface-level for some — mentions the concept without describing implementation. May confuse PSF controls with general software practices.

Insufficient0–9 pts

Fewer than 4 PSF domains addressed or significant misapplication of framework concepts. Evidence is entirely absent or plagiarised.

D3

Risk Assessment

20 pts

Identification, analysis, and mitigation of AI-specific risks in the deployment context.

Exemplary18–20 pts

Systematic risk register with likelihood × impact scoring. AI-specific risks (hallucination, prompt injection, model drift, third-party dependency) clearly distinguished from general software risks. Mitigations are proportionate and evidenced.

Proficient14–17 pts

Clear identification of major AI-specific risks with reasonable mitigations. Some risks may lack severity scoring. Mitigation evidence partially present.

Developing8–13 pts

Generic risk list that doesn't distinguish AI-specific risks. Mitigations are superficial ('we will monitor it'). No evidence of risk review process.

Insufficient0–7 pts

No meaningful risk analysis. Risks either not identified or treated identically to standard software risks. No evidence of safety thinking.

D4

Operational Maturity

15 pts

Evidence of sustainable AI operations: monitoring, runbooks, on-call, continuous improvement.

Exemplary14–15 pts

Complete observability stack documented (metrics, logs, traces, alerts). Runbooks for common failure modes. Post-incident reviews on file. Clear escalation path. Evidence of proactive quality monitoring.

Proficient11–13 pts

Core monitoring in place with some alerting. At least one post-incident review documented. Escalation path defined if not fully documented.

Developing6–10 pts

Basic error logging only. No structured incident response. On-call undefined. Monitoring described as planned rather than implemented.

Insufficient0–5 pts

No operational processes documented. System is effectively unmonitored. No evidence of structured response to failures.

D5

Communication Clarity

15 pts

Quality of portfolio documentation, technical writing, and ability to communicate AI risk to non-technical stakeholders.

Exemplary14–15 pts

Portfolio is clearly structured with executive summary, technical detail, and evidence appendix. Complex AI concepts explained without jargon where appropriate. Stakeholder communication examples included (e.g., incident comms, board updates).

Proficient11–13 pts

Well-structured portfolio. Technical writing is clear and mostly jargon-free. Some stakeholder communication evidence present.

Developing6–10 pts

Portfolio is organised but inconsistent in quality. Heavy on jargon. Minimal stakeholder communication evidence. Reader must infer context frequently.

Insufficient0–5 pts

Disorganised or incomplete documentation. Technical claims are unsubstantiated. Communication quality would impede professional use.

Assessment Process

1
Submit Expression of Interest
Complete the CPAP application form with your current role, employer, and brief description of your deployment project.
2
Portfolio Submission
Upload your portfolio (max 30 pages PDF or equivalent). Portfolio must include evidence for all 5 dimensions. Submissions open monthly.
3
Initial Screening
PAI staff screen for completeness and basic eligibility. Incomplete submissions are returned with feedback within 5 business days.
4
Expert Review
Two independent assessors score each dimension independently. Scores are reconciled; cases differing by >5 points trigger a third review.
5
Decision & Feedback
Pass (≥70/100) or Fail with detailed per-dimension feedback. Feedback is provided regardless of outcome.
6
Appeals
Failed candidates may appeal within 30 days. Appeals are reviewed by the PAI Standards Committee. One appeal permitted per submission.

Portfolio Guidance

What to include

  • Architecture diagrams with AI components labelled
  • Monitoring dashboard screenshots (redact PII)
  • Incident post-mortems or blameless retro notes
  • Risk register with mitigations
  • PSF compliance mapping table
  • Stakeholder communication examples

What not to include

  • Proprietary code or trade secrets
  • Unredacted customer PII
  • API keys, credentials, or connection strings
  • Content you did not personally create or lead
  • Marketing material or sales decks

Production AI Institute · CPAP Rubric v1.0 · Effective January 2026

Assessment enquiries: contact us via the contact form