Is the AWS Bedrock Academy free?

Yes. The full course is free and includes 25 modules, 52 lessons, 16 labs, and quiz-based progression with no sign-up required.

How is Infinity AWS Bedrock Academy different from AWS Skill Builder?

Unlike AWS Skill Builder which often gates advanced hands-on labs behind paid subscriptions or AWS Builder ID logins, Infinity Tech Stack's AWS Bedrock Academy is 100% free, ungated, and browser-first with zero sign-up required. It includes 25 modules, 52 lessons, 16 hands-on labs, and 125+ quiz checks covering the Converse API, Bedrock Guardrails, AgentCore, and multi-model evaluations without requiring active AWS account credentials.

Who is this course for?

Developers, architects, and platform teams building AI systems on AWS Bedrock from MVP to production scale.

What does this Bedrock course cover?

The academy covers runtime inference, prompt engineering, guardrails, knowledge bases, agents, evaluation, observability, security, compliance, resilience, incident response, and production launch discipline.

What should happen before promoting prompt changes?

Run quality and safety regressions — Regression testing is required to avoid silent behavior regressions.

Prompt Versioning and Promotion Tutorial | MLOps and PromptOps

← Back to Dashboard

1. Prompt Versioning and Promotion2. CI/CD Quality Gates for AI

Prompt Versioning and Promotion

📚 MLOps and PromptOps⏱ 10 min⭐ 110 XP

PromptOps as Engineering Discipline

Prompts, guardrails, and routing rules are release artifacts. Store them in version control with tests and changelogs - a one-word prompt edit can shift production behavior more than a thousand-line code change, yet most teams ship prompt edits with less review than a CSS tweak.

Version by Behavioral Impact

Bump	Example change	Required validation
Patch (3.2.1)	Typo fix, formatting - no behavioral intent	Smoke eval subset
Minor (3.3.0)	New instruction, added example, tone shift	Full golden + safety suites
Major (4.0.0)	Output schema change, role redefinition, model swap	Full suites + canary rollout + consumer coordination

Repository Layout That Works

prompts/
  ticket-triage/
    v3.2.0.md          # the prompt text, immutable once tagged
    changelog.md       # what changed and WHY ("reduced P1 over-escalation")
    evals/
      golden.jsonl     # task cases + expected properties
      scores-v3.2.0.json
routes/
  routing-matrix.yaml  # task → model tier, versioned alongside
guardrails/
  gr-support-x7.json   # exported config, reviewable in PRs

Promotion Flow

PR with prompt diff + changelog rationale - reviewed like code.
CI runs regression + safety suites against the candidate.
Promote through dev → staging → prod with the same artifact (no prod-only edits, ever).
On Bedrock, Prompt Management can hold versioned prompt resources your runtime references by version - promotion becomes a pointer update.

Semantic-version prompts by behavior impact.
Run regression and safety suites per change - no exceptions for "tiny" edits.
Promote through staged environments with the exact artifact you tested.

Anti-pattern: prompts living in a dashboard textbox, edited live in production. If the current prompt isn't reproducible from version control, you don't have a rollback - you have an archaeology project.

🧪 Knowledge Check

Press 1-4 to select1 of 2

PromptOps treats prompts as:

Temporary notes

Versioned deployment artifacts

Secrets

UI-only text