Executives and business leadersAmazon Bedrock for Leaders
Duration: Half day (proposed) · Format: Workshop, on site or live online
Who it is for
Sponsors deciding how generative AI is built on AWS.
Prerequisites
Familiarity with your organisation's AI ambitions. No technical background needed.
Learning outcomes
- Understand what Bedrock provides and what stays your responsibility
- Frame model choice, cost and risk trade-offs
- Set accountability for AI applications
Curriculum
- Bedrock capabilities in plain language
- Shared responsibility for AI workloads
- Use-case prioritisation
- Governance and funding model
Hands-on exercise
Assign owners and review points to a candidate application.
Takeaway materials
- Decision checklist for approving AI use cases
- One-page adoption roadmap template
- Board-level briefing outline
Completion outcome
ArniTech completion certificate (not a vendor certification).
Risk, compliance and security professionalsGuardrails and Access Boundaries on Amazon Bedrock
Duration: 1 day (proposed) · Format: Workshop with review exercises
Who it is for
Security, risk and compliance reviewers.
Prerequisites
Working knowledge of your organisation's risk, security or compliance processes. No coding needed.
Learning outcomes
- Evaluate what guardrails do and where they fall short
- Review IAM, network and data boundaries
- Define evidence for audit
Curriculum
- Guardrail types and their limits
- IAM and network boundaries
- Logging and model invocation records
- Testing for misuse and leakage
Hands-on exercise
Review a sample Bedrock design and list gaps with proposed controls.
Takeaway materials
- Control-mapping worksheet
- Review checklist for AI use cases
- Evidence examples to request from delivery teams
Completion outcome
ArniTech completion certificate (not a vendor certification).
Engineers, architects and technical teamsSecure RAG and Guardrails on Amazon Bedrock
Duration: 2 days (proposed) · Format: Hands-on lab course
Who it is for
Developers, data engineers and cloud architects.
Prerequisites
Comfortable with APIs, cloud fundamentals and at least one programming language. A sandbox account is arranged with you before the course.
Learning outcomes
- Build a retrieval-augmented application with permission-aware access
- Configure and test guardrails
- Evaluate quality and safety before release
Curriculum
- Foundation model selection
- Knowledge bases, chunking and retrieval quality
- Access control for source documents
- Guardrail configuration and testing
- Evaluation, monitoring and release gates
Hands-on exercise
Build a RAG prototype and attempt to make it reveal restricted content.
Takeaway materials
- Reference architecture notes
- Annotated code samples from the labs
- Evaluation and release checklist
Completion outcome
ArniTech completion certificate (not a vendor certification).
Business users and functional teamsUsing Bedrock-Powered Assistants at Work
Duration: Half day (proposed) · Format: Interactive workshop
Who it is for
Business teams using internal assistants built on Bedrock.
Prerequisites
None beyond access to the relevant tool (or a demonstration environment we provide).
Learning outcomes
- Ask effective questions of company-knowledge assistants
- Spot unreliable or out-of-scope answers
- Report problems through the right channel
Curriculum
- How these assistants get their answers
- Question and follow-up patterns
- Checking sources
- Acceptable use and escalation
Hands-on exercise
Test your assistant with real questions and log the failures.
Takeaway materials
- Prompt and task pattern library for your functions
- Do and don't guide for sensitive data
- Quick-reference card
Completion outcome
ArniTech completion certificate (not a vendor certification).
