Eight things we do on AWS
Most engagements start with a review or a migration and grow into running the platform. You can also hire us for one piece and stop there.
What each engagement covers
Each one is scoped and priced on its own. Take one, or take the lot.
AWS Well-Architected Review
A two-week read of your AWS account against the six pillars. You get a ranked list of what to fix, what it costs, and what happens if you leave it.
AWS migration
Data centre, another cloud, or a half-finished migration someone else started. We plan the cutover, rehearse it, and keep the rollback usable.
SaaS transformation
Turn a single-tenant product into a multi-tenant platform on AWS: tenant isolation, metering, self-serve onboarding and a per-tenant view of cost.
Multi-tenant web apps
Pool, silo and bridge tenancy models on AWS, with per-tenant limits, isolation you can prove to an auditor, and cost attribution that adds up.
Serverless environments
Lambda, API Gateway, EventBridge and Step Functions built where they pay off, and left alone where a container is simply cheaper to run.
DevOps and platform
Terraform, deployment pipelines and environments that rebuild from code. Deploys a new hire can run in their first week without supervision.
Modern app development
Containers on EKS or ECS, microservices where they earn their keep, and enough observability to debug the system at three in the morning.
Compliance and security
IAM cleanup, encryption, audit trails and the evidence collection that turns a SOC 2 or ISO 27001 questionnaire into a short afternoon.
Review, fix, run, trim
In that order. We do not take the pager for a system we have not read first.
- 01
Review
Week 1 to 2We read your account, not your slide deck. Cost breakdown, IAM exposure, single points of failure, and what breaks first at 3x traffic.
- 02
Fix
Week 3 onwardThe ranked list gets worked top down. Infrastructure moves into Terraform as we touch it, so the next change is repeatable.
- 03
Run
OngoingWe take the pager. Alarms, runbooks, patching, backups and restore drills that actually get tested.
- 04
Trim
Every quarterReserved capacity, right-sizing, storage tiers and a written report on where the money went and what we removed.
How to pick the right one
The eight above split into four jobs. Find out what is wrong, move something, build something new, or keep it running. Almost every enquiry we get is one of those, and the wrong starting point is usually the expensive one.
Find out what is wrong. The Well-Architected review is the two-week version: we read the account, the Terraform and the alarms, then hand back findings ranked by cost and risk. Pick this when the symptom is vague. A bill that grew 40% with no new customers, or a deploy nobody wants to run on a Friday, is a symptom, not a diagnosis.
Move something. Migration work covers datacentre to AWS, account to account, and one region to several. The estimate people want to hear is three months. On a regulated ledger it is not true, and we say so before the contract rather than in month four.
Build something new. Modern application work, serverless, multi-tenant SaaS and container platforms sit here. These start with a design document you can disagree with in writing, because disagreeing with a diagram costs an afternoon and disagreeing with a running platform costs a quarter.
Keep it running. DevOps and compliance work is the part that does not end: pipelines, guardrails, evidence for auditors, and the on-call rotation. We take this on for systems we have read. Not for ones we have not.
If two of those describe you, take the review first. It is the cheapest way to find out which of the other three you actually need, and its findings feed straight into the scope of whatever comes next.
Questions before you enquire
The ones that come up in almost every first call. Anything missing, email [email protected].
Which service should we start with?
If nobody can currently say what the account costs per team or when a restore was last tested, start with the Well-Architected review. It runs two weeks and ends in a written list of findings ranked by what they cost you. If you already know the problem and want it fixed, skip the review and hire us for that piece.
Do we have to take more than one?
No. Each engagement is scoped and priced on its own, and every one ends with the work in your repository and a handover document. Most clients add a second later because the first one surfaced something, not because the contract required it.
What does a fixed scope actually mean here?
The statement of work names the deliverables, the duration and the price before anyone starts. If the work turns out larger than we estimated, we say so and you decide whether to extend. We do not bill discovery as delivery.
Who does the work?
A named engineer who stays on your engagement start to finish, not a queue. You get their name in the kickoff and you talk to them directly, in your Slack or ours.
Can you run the platform after the project ends?
Yes, on a managed arrangement with an agreed response time. We only take the pager for a system we have read and, in most cases, helped build. Taking on-call for infrastructure nobody on our side has opened is how outages get answered slowly.
How long does each engagement take?
Two weeks for a Well-Architected review. Four to twelve weeks for most build work. Migrations depend entirely on what is being moved: the insurance ledger in our case studies took nine months with six engineers, and the first two of those months were discovery.
Do you work with teams that already have an AWS setup?
That is most of our work. Greenfield accounts are rare. The usual starting point is an account somebody built three years ago, that has since grown past what its original design assumed, and that nobody wants to touch.
Send us your AWS bill
Thirty minutes with an engineer, no slide deck. We will tell you what we would change first and what it would cost to have us do it.