Skip to content

Administration

Roles and permissions in Public Cloud Infrastructure

Each role in Public Cloud Infrastructure: what it sees, does and cannot do, its approvals, and a permissions matrix.

Applies to Public Cloud Infrastructure as it works today. Moraph updates the platform continuously, and this page is reviewed when it changes.

Last reviewed 5 min readReport a problem
On this page
  1. Roles
  2. Platform owner
  3. Security lead
  4. Application engineer
  5. Moraph cloud engineer
  6. Moraph security operations analyst
  7. Moraph service manager
  8. Auditor or examiner (read-only)
  9. Permissions at a glance
  10. How access is controlled

Who can see and do what in Public Cloud Infrastructure. These roles are the starting point; your administrators set them up to match your teams. Platform roles are assigned in AWS Financial Services as part of the managed service.

Roles

Platform owner

Head of infrastructure, cloud platform lead, IT director. Owns the institution's AWS environment and decides what is built, what changes and who has access.

  • Sees: All accounts, workloads and their compliance with the baseline; Open change requests, exceptions and findings; Spend by account and workload; Monthly service reports and the change record
  • Does: Approves new accounts and changes to the landing zone; Approves changes to production infrastructure under your change policy; Certifies who has access to which accounts and roles; Reviews the monthly service report with Moraph; Sets tagging and cost allocation rules
  • Cannot: Approve a guardrail exception alone; the security lead approves it as well; Edit or delete entries in the audit trail or change record
  • Approvals: Approves accounts, production changes and access certifications; exceptions also need the security lead.
  • Notified about: Change requests waiting for approval, failed deployments, escalated findings and access reviews coming due.

Security lead

CISO, information security officer, cloud security lead. Owns the guardrail baseline and the institution's decisions about cloud risk.

  • Sees: The guardrail baseline and its version history; All findings, their triage and their outcome; The exception register with review dates; Incident records and escalations
  • Does: Approves the baseline and changes to guardrails; Approves, renews or removes exceptions; Decides on escalated findings involving consumer data; Leads incidents and decides on notification; Sets which AWS regions and services are allowed
  • Cannot: Apply guardrail changes directly; they go through the pipeline; Approve an exception they requested themselves
  • Approvals: Approves guardrail changes and exceptions; their own requests need a second approver your organization names.
  • Notified about: High-severity findings, escalations, exceptions near their review date and guardrail changes ready for approval.

Application engineer

DevOps engineer, application developer, site reliability engineer. Builds and runs the institution's workloads in AWS within the guardrails.

  • Sees: Their team's accounts and the resources in them; Pipeline results, including which policy checks failed and why; Findings raised on their team's resources
  • Does: Requests new accounts through account vending; Deploys infrastructure as code through the pipeline; Fixes resources flagged against the baseline; Requests exceptions with a reason and compensating control
  • Cannot: Change service control policies, logging or security tooling; Make storage public or disable encryption where guardrails prevent it; Deploy to production without the approvals your change policy requires
  • Approvals: Production changes are approved by the platform owner; exception requests by the platform owner and the security lead.
  • Notified about: Failed policy checks, findings on their resources and decisions on their requests.

Moraph cloud engineer

Moraph cloud engineer, platform engineer. Builds and maintains the landing zone, guardrail code and pipeline, and applies approved changes.

  • Sees: Landing-zone and guardrail code and its history; Account configuration and drift across the organization; Change requests and their approvals; AWS service health and announced service changes
  • Does: Writes and tests guardrail and infrastructure changes; Applies approved changes through the pipeline; Enrolls new accounts and applies the baseline; Corrects drift under agreed runbooks; Opens and works cases with AWS Support on your behalf
  • Cannot: Approve their own changes; Access workload data in your accounts unless you grant access for a specific task; Grant exceptions
  • Approvals: Every change they make is approved by your organization, and reviewed by a second Moraph engineer before it is applied.
  • Notified about: Approved changes ready to apply, failed deployments, drift findings and AWS service notices.

Moraph security operations analyst

Moraph security operations analyst, SOC analyst. Watches findings across every account and handles or escalates each one.

  • Sees: Findings from AWS Config, GuardDuty and Security Hub; Account owners, escalation contacts and runbooks; Earlier findings and how they were resolved
  • Does: Triages findings and sets severity; Applies runbook corrections such as re-enabling S3 Block Public Access; Escalates findings that need your decision; Records the action taken and closes findings
  • Cannot: Accept risk or close a finding as an exception; Change guardrails; Read the contents of your data stores
  • Approvals: Corrections outside an agreed runbook need approval from your security lead.
  • Notified about: New findings, by severity, and responses to escalations.

Moraph service manager

Moraph service delivery manager. Runs the service relationship: reporting, reviews, priorities and anything that is not working.

  • Sees: Service reports, open requests and escalations; Exceptions and their review dates; Spend and usage trends
  • Does: Prepares and leads the monthly service review; Tracks requests, changes and improvements to completion; Coordinates with AWS on account and support matters; Manages escalations about the service itself
  • Cannot: Approve changes or exceptions; Change configuration in your accounts
  • Approvals: None of their own; they bring decisions to your platform owner and security lead.
  • Notified about: Requests past their target, escalations and items due for the next review.

Auditor or examiner (read-only)

Internal auditor, external auditor, regulatory examiner (under temporary access). Reviews cloud controls and evidence without being able to change anything.

  • Sees: The guardrail baseline and its history; Configuration compliance and drift for the period in scope; The change record, exception register and findings; Monthly service reports
  • Does: Reads and searches evidence within their scope; Exports evidence they have been given access to, with each export logged; Leaves review notes where your organization enables them
  • Cannot: Change, approve or close anything; See accounts or periods outside the scope granted
  • Approvals: Their access is approved by the platform owner or security lead, with a scope and an end date.
  • Notified about: Nothing by default.

Permissions at a glance

ActionPlatform ownerSecurity leadApp engineerMoraph engineerMoraph SecOpsService mgrAuditor
Request a new accountFullFullOwn work––––
Deploy infrastructure through the pipeline––Own workFull–––
Approve production changesApproves––––––
Propose a guardrail changeFullFull–FullFull––
Approve a guardrail changeApprovesApproves–––––
Apply guardrail changes–––Full–––
Request an exceptionFullFullOwn work––––
Approve an exceptionApprovesApproves–––––
Triage and correct findingsViewViewOwn workFullFullView–
Certify account accessApprovesView–––View–
View control evidenceViewViewOwn workViewViewViewView
View audit trailViewView–ViewViewViewView

"Own work" means only records or items assigned to that person; "Approves" means the role signs off the action.

How access is controlled

  • Nobody approves their own change or exception: Moraph engineers never approve the changes they write, and exceptions need both your platform owner and security lead.
  • Roles in AWS follow groups in your identity provider through IAM Identity Center, so joiners, movers and leavers are handled where you already manage them.
  • Every administrative action by Moraph staff is recorded in CloudTrail in a logging account they cannot alter.
  • Roles and approval routes are agreed during onboarding and can be scoped by account, environment or business line. Auditors and examiners get temporary read-only access with a scope and an end date.

Was this page helpful?

Report a problem with this page
What’s the problem?
Quote the step or heading if you can. Don’t include passwords or personal records.
Only if you’d like a reply. We use it for this report and nothing else.

Need help with your own setup rather than this page? Raise a support ticket.