Lambda
Memory and provisioned concurrency sized against real usage, not guesses.
Typical finding: old function versions still holding SnapStart snapshots.
Independent AWS cost reviews
I review your AWS account, find where the money goes, and give you a prioritised list of changes with estimated savings, each one checked against your code before it reaches you.
The services where spend quietly grows without anyone owning it. Each area comes with the kind of finding that usually turns up.
Memory and provisioned concurrency sized against real usage, not guesses.
Typical finding: old function versions still holding SnapStart snapshots.
Access patterns, write volume and indexes compared with what the application does.
Typical finding: Global Secondary Indexes nobody queries.
Every log source ranked by ingestion volume, with Step Functions and Lambda first.
Typical finding: verbose log levels recording full execution data on every state change.
Custom metrics in CloudWatch and Datadog checked for cardinality and actual use.
Typical finding: per-request tags that no dashboard or alert reads.
CPU and memory use across instances and clusters, plus backup retention.
Typical finding: clusters sized for a peak that no longer happens.
Usage data shows what looks wasteful. It doesn't show whether it's safe to change. An index with almost no traffic might serve a monthly report. A replica that looks idle might be your failover.
Before a finding goes in the report, I check it against your source and infrastructure code: who owns the resource, whether the live configuration matches what's defined in code, and whether a rarely used path still depends on it. You get fewer findings, and your team can act on all of them.
We talk about your setup, your spend and what you want to achieve.
You create an IAM role I can assume, protected by an External ID. I never handle long-lived credentials and can't change anything.
Usually under a week. I collect usage and cost data, then validate each finding against your code.
Every finding with its evidence and estimated saving, ranked so your team knows where to start.
I'm Higgor Leimig, a software engineer who specialises in AWS cost. I run the reviews myself, from the first call to the final report.
I'm a FinOps Certified Practitioner, certified by the FinOps Foundation.
Verify credential
An IAM role in your account that trusts mine, with an External ID and a read-only policy listing exactly which permissions it has. You review the policy before creating it, and you can delete the role when the review is done.
No. The role is read-only. Every change in the report is yours to make, when it suits your team.
Usually under a week from getting access, depending on the size of your estate.
It's how I check that a finding is safe to act on, so it makes the report more reliable. If that isn't possible, I'll mark which findings still need checking by your team.
Yes. I use Datadog metrics alongside CloudWatch to cross-check utilisation, and review Datadog custom metrics for cardinality.
Tell me a bit about your AWS setup and what you'd like to achieve. I'll reply within two working days.