Curriculum & Syllabus · CIWLEARNING
AI Operator
One program, ten modules, three stages. Each stage ends in its own credential, so a learner can stop at any stage and leave with something real.
The program in one page
AI Operator teaches people who are accountable for AI-enabled work to turn AI output into a bounded, reviewable, testable operation — and to prove whether it creates value. It is not a prompting course and not a model-building course.
The three stages answer three questions, in order. Can you govern a single AI workflow? Did it produce organizational value, and how would you know? Can you run it as an accountable operation over time?
| Stage | Modules | Time | Fee | Prerequisite |
|---|---|---|---|---|
| Core | 1 Scope · 2 Context · 3 Approval · 4 Warrant | 16–22 h | $100 | None |
| Strategic Track | 5 Strategic Alignment | 26–37 h incl. Core | $125 incl. Core | None |
| Operations | 6 Design · 7 Environment · 8 Execute · 9 Evaluate · 10 Deliver | 33–50 h | $165 | Strategic Track completion |
The work carries forward. The scope statements, context packages, approval matrices and records built in Core become the inputs Strategic measures and Operations runs. The product of this program is not the certificate — it is a chain of operating artifacts built around one real workflow the learner is responsible for.
Standards that apply to every stage
| Element | Standard |
|---|---|
| Knowledge checks | 80% to pass · unlimited attempts · no time limit |
| Summative assessments | 80% to pass · three attempts |
| Graded applications | Scored against a published four-level rubric: Emerging · Developing · Proficient (target) · Exemplary |
| Operator Journal | One reflective entry per module, graded |
| Discussion | One per module, scored against a rubric where all-Proficient equals 80% |
| Evidence | Every worked example in the curriculum comes from a real, dated operation, not a hypothetical one |
Stage one of three
Core — AI Operator
Can you govern a single AI workflow?
- Modules
- Four: Scope, Context, Approval, Warrant
- Time
- 16–22 hours
- Fee
- $100
- Prerequisite
- None
- Credential
- Certificate of Completion — AI Operator (Core)
Who this is for
Operations leaders, program and project managers, functional managers, consultants, founders and governance professionals who already have AI inside work they must answer for. Someone who wants better prompts does not need this course; someone who wants to build models does not either.
What you will be able to do
- State what a workflow may touch, what it may not, and what it must escalate.
- Keep specification, which binds, separate from context, which advises.
- Assemble the standing material an operation needs and test it on someone who does not do the work.
- Decide what evidence is enough before anything is accepted, and who may stop it.
- Write records that declare the conditions under which they stop being true.
Curriculum
| Module | Graded applications | Also |
|---|---|---|
| 1 · Scope What may this touch? |
1.1 Scope statement 1.2 Separate your documents 1.3 Veto design |
Teaching book · discussion 1.4 Journal 1.5 · knowledge check 1.6 (10 items) |
| 2 · Context What must it have in front of it? |
2.1 Build a context package 2.2 Objective to directive 2.3 Context decay check |
Teaching book · discussion 2.4 Journal 2.5 · knowledge check 2.6 (10 items) |
| 3 · Approval Who accepts this? |
3.1 Risk-tiered approval matrix 3.2 Verification ceiling 3.3 Veto map |
Teaching book · discussion 3.4 Journal 3.5 · knowledge check 3.6 (10 items) |
| 4 · Warrant Can this record still be acted upon? |
4.1 Warrant audit 4.2 Write a self-invalidating record 4.3 Enforcement test |
Teaching book · discussion 4.4 Journal 4.5 · knowledge check 4.6 (10 items) |
Artifacts you leave with
A scope statement (1.1), a context package (2.1), a risk-tiered approval matrix (3.1), a self-invalidating record (4.2) and an enforcement test (4.3). Together they govern one workflow end to end.
Assessment and credential
Twelve graded applications, four Operator Journal entries, four discussions and four knowledge checks, closing with the Core Summative Assessment covering Modules 1–4: twenty items, 80% to pass, three attempts.
Stage two of three
Strategic Track
Did it produce organizational value, and how would you know?
- Modules
- One: Strategic Alignment (Module 5)
- Time
- 26–37 hours in total, including Core
- Fee
- $125, includes Core
- Prerequisite
- None — Core is included
- Credential
- Certificate of Completion — AI Operator (Strategic Track). Requires both summatives
Who this is for
The operator who has governed a workflow and now has to defend it — to a sponsor, a budget holder, or a board. A governed process with no purpose is a well-run process with no reason to exist.
What you will be able to do
- State the problem before naming the tool, and recognize a solution looking for one.
- Propose measures that survive the free-to-fake test, and discard the ones that do not.
- Judge a tool on capability, boundability, auditability, attributability and exit cost — from observed evidence rather than vendor material.
- Bring security, compliance, legal, subject-matter experts and IT operations in with an artifact each can check.
- Make the strongest case against your own position, and say what would change your mind.
- Teach one discipline to a colleague and judge transfer rather than satisfaction.
Curriculum
| Component | Work |
|---|---|
| Graded applications | 5.1 Business case · 5.2 Measurement plan · 5.3 Argue against the instructor 5.4 Tool decision record · 5.5 Stakeholder alignment plan · 5.6 Teach an operator |
| Discussion | 5.7 ISO 42001 or NIST AI RMF? |
| Journal | 5.8 Operator Journal — Module 5 |
| Knowledge check | 5.9 Strategic Alignment — sixteen items |
| Strategic Lounge | An open forum for the cohort |
Artifacts you leave with
A business case (5.1), a measurement plan (5.2), a tool decision record (5.4) and a stakeholder alignment plan (5.5) — together, the case for whether the initiative deserves organizational support, including the decision to stop.
Assessment and credential
Six graded applications, one journal entry, one discussion and one knowledge check, closing with the Strategic Summative Assessment: thirteen items, 80% to pass, three attempts. The Strategic credential requires both the Core and Strategic summatives.
Stage three of three
Operations
Can you run it as an accountable operation over time?
- Modules
- Five: Design, Environment, Execute, Evaluate, Deliver
- Time
- 33–50 hours — about half the whole program
- Fee
- $165
- Prerequisite
- Strategic Track completion
- Credential
- Certificate of Completion — AI Operator (Operations), earned by passing the capstone at 80%
Who this is for
The operator moving from a governed workflow to a standing human–agent operation: several agents, named identities, a shared record, supervision that scales, and an ending decision somebody has to make.
What you will be able to do
- Design an operating concept another competent operator could use and challenge.
- Map a real environment by the identity that acts at each hop and the observation that establishes it — not the label the interface displays.
- Turn a prompt into an executable brief with authority stated by verb and stopping conditions that name events.
- Design a shared record in which a lost write arrives with its own name rather than as a retryable error.
- Calculate a span of control sized to the peak, treating human review as neither free nor elastic.
- Build controls that can fail honestly, and tell an acknowledgment from an authoritative readback.
- Assemble sponsor evidence that presents the option you dislike at full strength.
Curriculum
| Module | Graded applications | Also |
|---|---|---|
| 6 · Design | 6.1 Operating concept · 6.2 PEAR readiness 6.3 Authorized exception |
Book · discussion 6.4 · journal 6.5 Knowledge check 6.6 (15) |
| 7 · Environment | 7.1 Environment map · 7.2 Access-and-authority matrix 7.3 Secure bounded deployment |
Book · guided labs · discussion 7.4 Journal 7.5 · knowledge check 7.6 (17) |
| 8 · Execute | 8.1 Executable brief · 8.2 Shared-record protocol 8.3 Span of control |
Book · discussion 8.4 · journal 8.5 Knowledge check 8.6 (15) |
| 9 · Evaluate | 9.1 Test and control plan · 9.2 Supervision and dashboard 9.3 Regression and incident control |
Book · discussion 9.4 · journal 9.5 Knowledge check 9.6 (15) |
| 10 · Deliver | 10.1 Value-stream report · 10.2 Sponsor evidence pack 10.3 Capstone |
Book · discussion 10.4 · journal 10.5 Knowledge check 10.6 (15) |
The capstone
Operations closes with a performance assessment rather than a written test. The capstone runs in a bounded publishing lab reached through a browser terminal under the learner’s own identity. The learner declares opening orders before running anything, takes bearings at a declared interval, orders at least two corrections sized to the evidence, and absorbs one unannounced failure introduced by the instructor.
Scoring is against a four-criterion rubric. The learner is not graded on avoiding the failure — they are graded on whether their bearings detect it, whether the correction is proportionate, and whether the record shows who performed, approved, owned and could accept the residual risk.
Assessment and credential
Fifteen graded applications, five journal entries, five discussions and five knowledge checks, plus the Operations Summative Assessment covering Modules 6–10: fifteen items, 80% to pass, three attempts. The summative is a readiness check and does not gate the credential; the capstone does.