Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Course Outline
Foundations: Understanding the EU AI Act for Technical Teams
- Key obligations and terminology relevant to developers and operators.
- A technical perspective on prohibited practices outlined in Article 4.
- Mapping legal requirements to practical engineering controls.
Developing Secure and Compliant Software Lifecycles
- Structuring repositories and implementing policy-as-code for AI projects.
- Conducting code reviews and utilizing automated static checks to identify risky patterns.
- Managing dependencies and supply chains for model components.
Designing Compliance-Driven CI/CD Pipelines
- Defining pipeline stages: build, test, validation, packaging, and deployment.
- Integrating governance gates and automated policy checks.
- Ensuring artifact immutability and tracking provenance.
Model Testing, Validation, and Safety Verification
- Executing data validation and bias detection tests.
- Assessing performance, robustness, and resilience against adversarial attacks.
- Automating acceptance criteria and generating test reports.
Model Registry, Versioning, and Provenance Management
- Utilizing tools like MLflow for tracking model lineage and metadata.
- Versioning models and datasets to ensure reproducibility.
- Recording provenance details to produce audit-ready artifacts.
Implementing Runtime Controls, Monitoring, and Observability
- Instrumenting systems to log inputs, outputs, and decision-making processes.
- Monitoring for model drift, data drift, and performance metrics.
- Configuring alerting mechanisms, automated rollback procedures, and canary deployments.
Ensuring Security, Access Control, and Data Protection
- Applying least-privilege IAM policies for model training and serving environments.
- Protecting training and inference data both at rest and in transit.
- Managing secrets and adhering to secure configuration practices.
Enhancing Auditability and Evidence Collection
- Generating machine-readable logs and human-readable summaries.
- Packaging evidence effectively for conformity assessments and audits.
- Establishing retention policies and ensuring secure storage of compliance artifacts.
Managing Incident Response, Reporting, and Remediation
- Detecting suspected prohibited practices or safety incidents.
- Executing technical steps for containment, rollback, and mitigation.
- Preparing technical reports for governance bodies and regulators.
Summary and Future Steps
Requirements
- A solid understanding of software development and deployment workflows.
- Experience with containerization and foundational Kubernetes concepts.
- Familiarity with Git-based source control and CI/CD practices.
Target Audience
- Developers responsible for building or maintaining AI components.
- DevOps and platform engineers tasked with deployment operations.
- Administrators overseeing infrastructure and runtime environments.
14 Hours
