Skip to main content

Opportunities and Deep Dives

Cloud ex Machina turns cloud findings into opportunities: actionable items that can reduce cost, improve allocation, or strengthen operational posture. Dex helps users understand, discuss, and move those opportunities forward.

What an opportunity includes

An opportunity typically brings together:

  • The affected asset, service, account, project, or workload.
  • The reason Cloud ex Machina flagged it.
  • Estimated cost impact, when available.
  • The current state and suggested direction.
  • Ownership and routing context.
  • Review status and follow-up activity.

The goal is to make the opportunity actionable, not just visible.

How Dex helps with opportunities

Dex can explain an opportunity in the level of detail needed for the audience.

Ask Dex to:

  • Summarize the opportunity for an executive, FinOps lead, or engineer.
  • Explain the evidence behind the recommendation.
  • Identify likely owners or reviewers.
  • Compare possible actions.
  • Call out risks, assumptions, and unknowns.
  • Draft a message or ticket description.
  • Prepare next steps for review.

Dex can also continue the discussion as users ask follow-up questions, challenge assumptions, or add context.

Artifacts for collaboration

When an opportunity needs shared review, ask Dex to generate an artifact. Artifacts give the team a common surface for analysis, plans, and proposed code changes instead of keeping everything inside a chat thread.

Common opportunity artifacts include:

  • Deep-dive analysis.
  • Remediation plans.
  • Owner handoff notes.
  • Risk and decision summaries.
  • Tables and graphs for cost, savings, or option comparisons.
  • AWS, Azure, or GCP architecture diagrams.
  • Mermaid diagrams for flows, dependencies, and decision paths.
  • Code-change proposals or generated code changes for review.
  • Pull request summaries that explain the recommended change.

Users can iterate on an artifact in two ways:

  • Continue the discussion with Dex and ask for a revised version.
  • Add inline comments to specific parts of the artifact, then send those comments to Dex to produce an updated version.

This is useful when reviewers agree with the overall direction but want changes to wording, assumptions, scope, sequencing, or implementation details.

When artifacts include visualizations, users can download diagrams and graphs as images. Tables can also be downloaded as CSV files so teams can reuse the underlying data in spreadsheets, tickets, or presentations.

Deep dives

A deep dive is an evidence-backed investigation of an opportunity. It is meant to help a team decide whether the opportunity is worth acting on and how to proceed.

A useful deep dive should clarify:

  • What is being recommended.
  • Why the recommendation exists.
  • Which assets, owners, or workloads are involved.
  • What cost impact is expected.
  • What operational risk should be reviewed.
  • What information is still uncertain.
  • Which next action is most practical.

For complex opportunities, ask Dex to include visuals in the deep dive. A table can compare options side by side, a chart can show cost or usage trends, and an architecture diagram can make the affected resources easier to understand.

Deep dives are decision support. They do not replace engineering review, approval, testing, or change management.

From discovery to action

Opportunity work usually follows this path:

  1. Cloud ex Machina surfaces a potential improvement.
  2. Dex helps explain the opportunity and gather context.
  3. Dex generates an artifact when the team needs a shared analysis, plan, or code-change draft.
  4. The team reviews the evidence, risk, and owner.
  5. Reviewers discuss with Dex or leave inline comments on the artifact.
  6. Dex incorporates feedback into an updated version when requested.
  7. Dex helps prepare the handoff, ticket, message, remediation plan, code diff, or pull request.
  8. The responsible team decides whether and how to act.
  9. Follow-up work is tracked until the opportunity is resolved, deferred, or dismissed.

Not every opportunity should be implemented immediately. Some need more data, some need owner review, and some may be intentionally deferred because the savings do not justify the risk or effort.

Working with owners

Dex can help make owner handoff clearer. A good handoff should include:

  • The resource or workload involved.
  • The expected benefit.
  • Why the owner is being asked to review it.
  • The decision needed.
  • Any known risk, timing, or dependency.
  • A suggested next step.

Example request:

Draft a short message to the payments team explaining this opportunity, why it matters, and what decision we need from them.

Remediation planning

When code or ticketing integrations are connected, Cloud ex Machina can help prepare implementation work. Dex can help frame the change, summarize the expected outcome, and keep discussion tied to the original opportunity.

For remediation work, artifacts are useful for reviewing the plan before action. A team can ask Dex to generate a plan or code-change draft, comment on the parts that need adjustment, and then ask Dex for a new version that reflects the review.

When Git repositories are connected, Dex can also help shift the work left into the systems engineers already use. It can identify likely code locations, make specific recommendations, generate code diffs, and open pull requests when your team's integration settings allow it.

Before applying any change, review it through your normal engineering process. Confirm the affected environment, test requirements, rollback plan, and owner approval.

When to ask for a deep dive

Ask Dex for a deep dive when:

  • The savings look meaningful but the action is not obvious.
  • The recommendation affects a production or customer-facing workload.
  • Ownership is unclear.
  • The team needs evidence before opening a ticket.
  • There are multiple possible actions and you need a comparison.
  • You need a concise summary for an approver.

Use lighter-weight questions for simple quick wins, such as unused resources in non-production environments or straightforward cleanup tasks.