Get in Touch
 Duration 14 hours

Course Outline

Foundations: The EU AI Act for Technical Teams

  • Relevant obligations and key terminology for developers and operators.
  • A technical perspective on understanding prohibited practices under Article 4.
  • Translating legal requirements into actionable engineering controls.

Secure and Compliant Development Lifecycle

  • Structuring repositories and implementing policy-as-code for AI projects.
  • Conducting code reviews and applying automated static checks for risky patterns.
  • Managing dependencies and supply-chain integrity for model components.

CI/CD Pipeline Design for Compliance

  • Defining pipeline stages: build, test, validation, package, and deploy.
  • Integrating governance gates and automated policy verification.
  • Ensuring artifact immutability and maintaining provenance tracking.

Model Testing, Validation, and Safety Checks

  • Implementing data validation and bias detection tests.
  • Assessing performance, robustness, and adversarial resilience.
  • Establishing automated acceptance criteria and generating test reports.

Model Registry, Versioning, and Provenance

  • Utilizing MLflow or equivalent tools for model lineage and metadata management.
  • Versioning models and datasets to ensure reproducibility.
  • Recording provenance data and generating audit-ready artifacts.

Runtime Controls, Monitoring, and Observability

  • Instrumenting systems to log inputs, outputs, and decision-making processes.
  • Monitoring model drift, data drift, and key performance metrics.
  • Implementing alerting systems, automated rollback, and canary deployments.

Security, Access Control, and Data Protection

  • Enforcing least-privilege IAM for model training and serving environments.
  • Safeguarding training and inference data both at rest and in transit.
  • Applying secrets management and secure configuration best practices.

Auditability and Evidence Collection

  • Generating machine-readable logs alongside human-readable summaries.
  • Packaging evidence for conformity assessments and regulatory audits.
  • Defining retention policies and ensuring secure storage of compliance artifacts.

Incident Response, Reporting, and Remediation

  • Identifying suspected prohibited practices or safety incidents.
  • Executing technical steps for containment, rollback, and mitigation.
  • Drafting technical reports for governance bodies and regulators.

Summary and Next Steps

Requirements

  • A solid grasp of software development and deployment lifecycles.
  • Proficiency with containerization and foundational Kubernetes concepts.
  • Working knowledge of Git-based source control and CI/CD methodologies.

Target Audience

  • Developers creating or maintaining AI-based components.
  • DevOps and platform engineers overseeing deployment processes.
  • Administrators responsible for managing infrastructure and runtime environments.

Number of participants


Price per participant

Upcoming Courses

Related Categories