Back to Certifications
NewFreeTrack-based

Get certified in

AI Work Competency Certification

Everyone says they use AI. This proves you use it well - that you can tell a real finding from a confident one, know what to check before you sign your name to it, and can do the work when the output is wrong.

The assessment

20 questions

per attempt

35 minutes

server-timed

75% to pass

weighted overall

60% Part C floor

plus no critical RAI failure

Part A
20%
Part B
30%
Part C
50%

Three tracks, one standard

Parts A and B are identical whichever track you pick. Only the four Part C work tasks change - so the certificate means the same thing for everyone, on the work they actually do.

Data Analyst

Datasets, SQL, dashboards and findings you have to defend.

Software Engineer

Generated code, tests, migrations and review decisions.

AI Engineer

RAG, evals, structured output, agents and safeguards.

What the 35 minutes look like

Every attempt draws a fresh paper - but always the same number of questions from each competency area, so nobody gets an easier test.

Part A20%

AI Knowledge

What you know about how AI systems behave and where they fail.

9 questionsfrom a bank of 20~7 min
  • AI fundamentals3
  • Prompting & context2
  • Verification2
  • Responsible AI1
  • Tools & workflows1
Part B30%

Situational Judgment

What you would actually do when AI output meets a real workplace situation.

7 questionsfrom a bank of 20~10 min
  • Validation & judgment2
  • Validation & judgmentwritten1
  • Privacy & responsibility1
  • Iteration & recoverywritten1
  • Delegationwritten1
  • Workplace pressurewritten1
Part C50%

AI at Work

Four written mini work simulations for your track.

4 questionsfrom a bank of 8 per track~18 min
  • AI output errorwritten1
  • Actual workflowwritten1
  • Validationwritten1
  • Judgment & recoverywritten1

You must score at least 60% here, whatever your overall total.

How the written answers are graded

A model reading an essay and returning “8.7 out of 10” is not a grade - it is a guess with a decimal point. So we split the job in two: the AI only observes, and the code scores.

The evaluator reports which specific behaviours it can evidence in your own words. It has no score field to write into. Points then come from a fixed rubric in code, so the same answer always earns the same mark - and the rubric is reviewable, not a black box.

1

Your answer

Written in your own words, in the time you have.

2

AI evaluator

Reports only which expected behaviours are evidenced, and which red flags appear.

3

Deterministic rubric

Code turns those observations into points, using the bank's own signal list.

4

Weighted result

Part A 20%, Part B 30%, Part C 50% - then the 75% bar, the Part C floor and the responsible-AI check.

Distinction

93% and above

Strong

85% and above

Certified

75% and above

Weekly Ranking

Frequently asked questions

Free, 35 minutes, and a verifiable certificate at the end.