Anonymised under NDA
Cloud cost optimisation agents
Cross-industry · A multi-cloud enterprise estate
FinOps agents that detect waste, propose remediations, and execute pre-approved changes under guardrails across AWS, Azure, and GCP.
The problem
- Cloud spend was growing faster than revenue with no clear single owner.
- Optimisation recommendations from native tooling were ignored because they lacked context.
- Manual remediation could not keep pace with the rate of new workloads.
The architecture
- Discovery agents correlate billing data with workload owners, tags, and recent deployments.
- Recommendation agents propose remediations with explicit cost / risk trade-offs.
- Execution agents apply pre-approved low-risk changes; everything else is routed to a human approver via chat ops.
AWSAzureGCPTerraformLangGraphSlack
The outcome
- FinOps moved from a quarterly review to a continuous, agent-led practice.
- Engineering teams received recommendations in their own language and chat tools, with full reasoning.
Services drawn on
A similar problem on your roadmap?
A Discovery Sprint is the fastest path to a scoped, costed proposal.