services

One service.
Measured in dollars saved.

Koali does one thing: we find waste in your AWS account and execute the changes to eliminate it. No re-architectures. No code rewrites. Just a methodical pass through your infrastructure with a savings guarantee on the other end.

what we do

A full-service optimization pass.

Most cost-optimization firms deliver a slide deck and call it done. The recommendations sit in a backlog. The savings never land. The retainer keeps running.

We work the other way around. We audit your account, walk you through exactly what we'd change and the dollar impact of each item, then — once you approve — we execute the changes ourselves, inside your change windows, with rollback paths ready. Savings show up in your next bill, not a quarterly status update.

Everything we do stays at the infrastructure layer. We don't rewrite your application, re-architect your systems, or tell you to switch databases. Those projects have their place — this isn't it.

the levers

Where the savings actually come from.

Every AWS account has its own shape of waste. Here's the surface area we check, and the levers we pull when we find something.

EC2 right-sizing

Match instance types and sizes to actual utilization. Most fleets are over-provisioned by 40%+.

Graviton migration

Move compatible workloads to ARM-based Graviton instances for 20–40% better price-performance.

Savings Plans & RIs

Optimize commitment coverage across compute and database. Convert, exchange, and layer terms intelligently.

EDP negotiation

Negotiate Enterprise Discount Programs and Private Pricing Agreements with AWS on your behalf.

RDS right-sizing

Tune RDS instance classes, storage types, and IOPS provisioning to match your actual workload.

Aurora optimization

Serverless v2 sizing, I/O-Optimized evaluation, and reader fleet right-sizing for Aurora clusters.

Lambda tuning

Memory optimization, ARM migration, and provisioned concurrency analysis for Lambda workloads.

S3 storage tiering

Lifecycle policies, Intelligent-Tiering, and Glacier transitions for cold data. Delete what shouldn't be there.

Data transfer

Cross-AZ, cross-region, and NAT traffic patterns. Often the most invisible and expensive line item.

CloudWatch retention

Log group retention policies, metric filter cleanup, and high-cardinality metric audits.

NAT Gateway alternatives

VPC endpoints and gateway routing to cut expensive NAT Gateway data processing charges.

Spot & idle cleanup

Spot instance adoption for fault-tolerant workloads. Unused EBS volumes, EIPs, snapshots, and load balancers.

how an engagement works

From kickoff to first savings.

Timelines scale with your environment. Smaller accounts can see realized savings in as little as a week. Larger fleets with tight change windows take longer — but we'll tell you exactly where we are at every step.

Phase 01

Days 1–3

Read-only audit

You grant us a read-only IAM role. We analyze spend patterns, commitment coverage, utilization, and architecture against every optimization lever. No writes, no changes, no risk.

Complete waste inventory with dollar impact per finding

Phase 02

Days 3–14

Plan & approval

We walk you through every recommended change — the savings, the risk, the rollback path. You approve what moves forward. Nothing happens without your signoff. Larger environments take the full two weeks; smaller ones can finish in a few days.

Itemized change plan with risk profiles and rollback paths

Phase 03

Your change windows

Execution

We escalate to write access and execute the approved changes inside your specified windows — overnight, weekends, whenever works. If you use Terraform, Pulumi, or CloudFormation, we push changes through your IaC so nothing drifts or gets reverted. If you don't, we make direct console changes with full documentation.

Executed changes with commit history and rollback readiness

Phase 04

Ongoing

Prove & monitor

We track realized savings against your Cost Explorer baseline, watch for anomalies, and only invoice once savings have actually shown up in your bill. If something drifts back, we know.

Monthly savings report tied to Cost Explorer data

Access model: read-only first, always. Write access is only granted after you've approved a plan — and it's scoped to exactly what we need to execute it.

what we don't do

Honest about the scope.

Knowing what we won't touch is as important as knowing what we will. If you need any of these, we'll happily point you somewhere good.

Application code changes

We don't touch your codebase. Optimization happens at the infrastructure layer.

System re-architecture

We won't tell you to rewrite services, switch databases, or adopt a new framework.

GCP, Azure, or other clouds

AWS only. We go deep on one cloud instead of shallow on three.

Vendor comparisons or RFPs

We're not a consultancy. We execute on AWS — that's the whole pitch.

tooling

We meet your infra where it lives.

If you run infrastructure as code, we push changes through your toolchain so nothing drifts or gets reverted on the next deploy. If you don't, we work directly in the console and document everything.

Terraform

HCL modules, workspaces, state management

Pulumi

TypeScript, Python, Go stacks

CloudFormation

Stacks, StackSets, SAM

Non-IaC

Direct console changes with full documentation

questions

What engineering leaders ask us first.

What size AWS accounts do you work with?

We work with companies spending $10,000/month or more on AWS. Below that, the shared-savings model doesn't make sense for either of us.

How is access handled?

Read-only IAM role first — always. We audit without the ability to change anything. Write access is only granted after you've approved a specific plan, and it's scoped to exactly what we need.

How fast will we see savings?

Small environments can see realized savings in as little as 7 days. Larger fleets with tight change windows take longer — the constraint is usually how much change time you give us, not the work itself.

How do you measure savings?

Against your Cost Explorer baseline, month over month, adjusted for usage growth. We invoice from realized savings only — not projections.

What if a change causes an issue?

Every change we make ships with a rollback path. If something behaves unexpectedly, we revert immediately. We also only execute inside your specified change windows so you're never surprised.

Who actually does the work?

Engagements are handled directly by Derek, Koali's founder — a decade of AWS experience and an AWS Certified Solutions Architect. You get a single technical point of contact, not a rotating bench.

ready when you are

Start with a free audit.

Read-only access, a dollar-backed findings report, and zero obligation. The only thing you risk is learning how much you've been overpaying.

Get a free audit

$10k+ monthly AWS spend · No credit card · No retainer