Description

A focused course, tailored for you

AI Governance Engineering: Model Risk to Audit Sign-off

Build the compliance documentation stack that enterprise customers and regulators actually accept.

The compliance request arrives after the model ships. A risk tier classification, a model card with explainability evidence, monitoring SLAs tied to the contract. The feature is technically correct. The governance documentation is not there. That reactive documentation cycle, repeated across every deployment, is the exact problem this course is built to eliminate.

$199 one-time

Tailored to your situation. Access within 24 hours. 30-day money-back.

Includes a hand-built implementation playbook delivered alongside course access, generated for your specific situation.

Why this course

Enterprise ML deployed at scale runs inside a governance framework built by risk teams, legal teams, and customer compliance officers. Those teams review every AI feature against their own audit checklist: EU AI Act risk tier, model card with explainability evidence, bias audit report, monitoring SLA tied to contractual commitments. The model's accuracy does not appear on that checklist. The documentation either passes or it does not. Most ML engineers write governance documentation after a compliance escalation, not before deployment, because the governance build was never part of their engineering process. The result is reactive work that delays deployments, strains customer relationships, and creates liability exposure on every model that shipped without a complete governance package.

What you walk away with

Classify any ML system against EU AI Act risk tiers and produce the classification rationale document your legal team can file.
Write a model card that satisfies both internal legal review and the external compliance audit your enterprise customers run.
Produce a bias and fairness audit evidence package that meets the requirements of financial services, healthcare, and public sector reviewers.
Build the monitoring governance document that ties your technical thresholds directly to the SLA obligations in your customer contracts.
Assemble the complete customer audit evidence file for any deployed AI feature, proactively, before the audit request arrives.

The 12 modules

Module 1. The Enterprise AI Compliance Landscape

An ML engineer shipping features to enterprise customers operates inside overlapping regulatory frameworks: EU AI Act risk tiers, NIST AI RMF profiles, customer-specific audit requirements, and contractual SLA obligations. This module maps the full landscape. You will identify which regulatory bodies have jurisdiction over your AI features, how customer industries add compliance layers on top of baseline regulations, and where your documentation obligations begin before a single model ships to production.

Module 2. Risk Tier Classification for ML Systems

The EU AI Act places ML systems in four risk tiers: unacceptable, high, limited, and minimal. Tier determines documentation burden, conformity assessment requirements, and post-market surveillance obligations. This module walks through the classification methodology, the prohibited-use categories relevant to enterprise platforms, and the high-risk categories that cover workflow automation, HR decision support, customer scoring, and infrastructure management. You will classify three representative systems and document the classification rationale.

Module 3. Writing Model Cards That Pass Legal Review

Most model cards engineers write describe the model. Legal and compliance reviewers want different content: intended use scope, documented out-of-scope uses, evaluation data provenance, fairness metrics across protected groups, and the specific limitations that bound the liability claim. This module provides the field-by-field template for model cards that pass legal review, with annotated examples for a customer-scoring model and a workflow recommendation engine deployed into enterprise environments.

Module 4. EU AI Act Conformity Assessment

High-risk AI systems require a conformity assessment before market deployment under the EU AI Act. This module covers the full assessment process: technical documentation requirements, quality management system obligations, the conformity declaration format, and post-market surveillance planning. You will build the technical documentation package for a sample enterprise ML feature, including the conformity declaration template your legal team can review and submit to the relevant national authority.

Module 5. NIST AI RMF Implementation for Platform Teams

NIST AI RMF organises AI governance into four functions: Map, Measure, Manage, and Govern. For a platform team deploying models to enterprise customers, each function produces specific artefacts: a risk profile, a measurement plan, a management policy, and a governance framework document. This module translates each AI RMF function into the concrete deliverables your team can build, maintain, and present to customers during vendor assessments and contractual audit cycles.

Module 6. Bias and Fairness Audit Evidence

Customers in financial services, healthcare, and public sector require bias audit evidence as a condition of deployment. This module covers the full audit methodology: protected-attribute identification, disparate-impact analysis, counterfactual fairness testing, and the audit report format compliance teams accept. You will work through a bias audit for a binary classification model, producing the evidence package that satisfies a customer compliance review and can be filed in a long-term audit evidence record.

Module 7. Explainability Documentation for Non-Technical Reviewers

Explainability is a documented claim with a methodology behind it, not just a technical output. This module covers SHAP summary reports, LIME explanations for individual predictions, and counterfactual explanation documentation. It also covers how to translate model-level explainability into the plain-language disclosure non-technical reviewers require. You will produce three explanation formats for the same model: one for the technical audit file, one for legal review, and one for an executive summary.

Module 8. Monitoring Governance and SLA Documentation

Enterprise ML contracts include monitoring obligations. This module covers the full monitoring governance stack: data drift detection thresholds, model performance degradation triggers, alerting SLAs, and the monthly compliance report customers expect at contract review. You will build the monitoring governance document that ties your technical thresholds to contractual commitments, including the escalation path, the SLA breach notification process, and the evidence format the customer audit team requires.

Module 9. Incident Response for AI Systems

When an AI system produces a harmful output, a biased decision, or an SLA breach, the response involves more than a code rollback. Under the EU AI Act, serious incidents require regulatory notification. This module covers the full incident response plan for AI systems: detection and classification, containment, regulatory notification obligations, customer communication protocol, the post-incident report format, and the corrective action evidence that formally closes the incident file.

Module 10. Building the Customer Audit Evidence File

Enterprise customers audit AI features at contract renewal, during procurement review, and on regulatory demand. This module walks through what a typical AI audit request includes and how to build the evidence file before the request arrives. You will assemble a complete customer audit evidence package: model card, risk tier decision with rationale, bias evidence, monitoring history, incident log, and the signed conformity declaration your legal team approved before deployment.

Module 11. Internal AI Governance Process Design

Consistent governance documentation across model deployments requires a repeatable internal process. This module covers governance process design for a platform engineering team: the model intake review checklist, the documentation sign-off workflow, the artefact repository structure, and the version-control discipline for model cards that stay accurate as models evolve. You will design the internal governance process for your team, including role assignments, review cadence, and the tooling that supports it across the deployment lifecycle.

Module 12. Sustaining Compliance Across Model Versions

Models retrain. Features update. Customer contracts renew with revised audit obligations. Regulations evolve. This module covers the lifecycle governance process: the triggers that require a documentation review, the version-control approach for compliance artefacts, and the continuous compliance posture that keeps every deployment audit-ready without rebuilding documentation from scratch on each release cycle. You will design the documentation refresh workflow and the tracking mechanism your team uses going forward.

How this addresses your situation

Specific modules that map to what you said you are dealing with.

Customer compliance team flags a shipped model for missing risk tier documentation: Modules 2 and 4 cover classification methodology and the conformity assessment package.

Legal review returns the model card as incomplete before a customer contract can execute: Module 3 covers the field-by-field template and annotated examples.

Enterprise customer requests bias audit evidence as a procurement condition: Module 6 walks through the full audit methodology and evidence package.

Contract renewal requires a monitoring governance document with SLAs tied to specific thresholds: Module 8 covers the monitoring governance build from threshold to contractual commitment.

What you get with this course

12 written modules covering AI governance documentation from risk tier classification to audit sign-off
Downloadable model card template with annotated field-by-field guidance for enterprise deployments
EU AI Act conformity assessment package template with a worked example for a high-risk system
Bias and fairness audit evidence template with a completed example for a binary classification model
Monitoring governance document template tied to a standard enterprise SLA structure
Customer audit evidence file template covering all required components for a full evidence package
Internal governance process design guide calibrated to a platform engineering team structure
Hand-built implementation playbook covering the specific feature type and regulatory context relevant to your role

What you will have in hand by Day 1, Week 1, Month 1

Within 24 hours your account in the learning environment is provisioned and the tailored implementation playbook is delivered alongside it.

All 12 modules and downloadable templates are immediately accessible on enrolment.

No scheduled sessions or live calls. Work through the material at your own pace alongside active deployment work.

Before and after

Before

Compliance escalation arrives after the feature ships. Three weeks of reactive documentation work follows. The audit delays the contract renewal and the compliance team questions whether governance is embedded in the build process at all.

After

Every deployment ships with a complete governance package. Risk tier classification, model card, bias evidence, monitoring plan. The compliance review is a formality. The customer audit evidence file exists before the request arrives.

What happens if you do not address this

Every deployment without a governance package is a compliance escalation waiting to happen. One customer audit that surfaces a missing model card or an undocumented risk tier can delay a contract renewal, trigger a regulatory enquiry, or produce a correction notice. The documentation gap compounds with every model that ships without it.

Who it is for

An ML engineer building enterprise features on a cloud platform. Technically proficient, shipping models into production workflows used by large organizations in regulated industries. Not running a research lab. Running a product that gets audited by the customers who buy it. Accountable for features that touch HR workflows, financial decisions, infrastructure automation, or customer-facing scoring, where enterprise procurement teams and regulators have specific documentation requirements that sit entirely outside the model training and deployment pipeline.

Who this is NOT for. ML researchers focused on academic publication without external audit obligations. Data scientists doing internal analytics where no customer compliance review applies. Engineers whose ML systems carry no regulatory classification, no contractual monitoring requirements, and no enterprise customer audit obligations.

How it arrives

Text-based course in the Art of Service learning environment, plus downloadable templates and worked examples for every module, plus the hand-built implementation playbook delivered alongside course access.

Time investment. 12 focused modules. Most engineers complete the core governance documentation for a single deployment within the first four modules. Full completion across all 12 modules and templates typically takes six to eight focused working sessions, each structured around a concrete deliverable.

Why $199 is the right number

Reading the EU AI Act and NIST AI RMF documentation directly tells you what is required but not how to build the artefacts. Internal legal or compliance teams advise on requirements but cannot produce the technical documentation on your behalf. Engaging a compliance consultant for a single model review costs ten to fifteen times the course price, without the transferable template library and repeatable internal process design you carry out of this course and apply to every subsequent deployment.

FAQ

Is this course specific to a particular regulatory region?

The core frameworks covered are EU AI Act and NIST AI RMF, both of which apply to enterprise ML deployed globally. Module 4 focuses on EU conformity assessment specifically. Modules 1 and 5 address both frameworks in parallel so you can apply whichever is relevant to your customer base and deployment geography.

Do I need a compliance background to take this course?

No. The course is built for ML engineers with a technical background who need to develop the governance documentation skills that sit alongside their engineering work. No prior compliance or legal experience is required. The course translates regulatory requirements into the specific artefacts you build.

What makes this different from reading the regulatory documentation directly?

The regulatory documents define what is required. This course walks you through building the artefacts: the model card template, the bias audit methodology, the monitoring governance document, the conformity assessment package. Each module ends with a deliverable you can apply to a current deployment.

How quickly can I apply what I learn to an active deployment?

Each module includes a worked example and a downloadable template calibrated to enterprise platform deployments. Most engineers complete the governance documentation for a single deployment in the first four modules. The implementation playbook provided on enrolment is built specifically for your role and feature context.

30-day money-back guarantee. If after a week of working through the materials this is not what you needed, reply to the receipt email and a full refund is processed. No questions, no forms.

Within 24 hours your account in the learning environment is provisioned and the tailored implementation playbook is delivered alongside it.