- Market
- Saudi Arabia
- Written for
- CIO, CTO
- Decision
- End configuration drift and make infrastructure reproducible
- Languages
- EN · AR
Information security for cloud delivery

ISO/IEC 27001
ISO/IEC 27001 certified
How drift happens
The small manual change that unravels the code
Drift is rarely a decision; it is a series of reasonable exceptions, and each has a discipline that prevents it rather than a tool that detects it after the fact.
An incident at 2am
The fix stays, and so does the code, because the process expected the emergency.
A fast path to make an emergency change and then bring it back into code the next morning, rather than a rule people break under pressure.
A quick fix in the console
No change without a trace, which is the only state in which code stays true.
Console write access removed for humans in the accounts that matter, so a quick fix has to go through the pipeline that records it.
A change made in a hurry with no review
A reviewed diff rather than a surprise, because the plan was legible.
A plan step that shows exactly what will change before it applies, small enough that a reviewer can actually read it.
State that no longer matches reality
The gap found on a Tuesday report, not during an outage.
Scheduled drift detection that reports the gap between code and the live estate before it becomes a failed apply.
The half nobody sells
Terraform is a language; the discipline is the product
Any competent engineer can express your cloud in Terraform. What decides whether the investment holds is what happens the first time someone is tempted to click "edit" in the console under pressure. If that path is open, the code drifts from reality within weeks and becomes a misleading map. If it is closed — enforced, not merely requested — the code stays the single source of truth it was meant to be.
So most of an honest engagement is organisational as much as technical: removing standing human write access to the accounts that matter, giving people a fast, recorded path for the genuine emergency, and reviewing infrastructure changes the way code changes are reviewed. The Terraform is the easy deliverable; the closed console is the one that pays back.
What we bring under code
The order we put an estate under code
Not everything at once, and not alphabetically — highest risk and highest change-rate first, so the value shows early.
1 · Identity and access, first — the most expensive place to make a mistake
- Accounts
- Roles
- Policies
- Trust relationships
2 · Network and boundaries — what a security review reads first
- VPCs
- Subnets
- Security groups
- Routing
3 · Data stores — residency and encryption described, not remembered
- Databases
- Object storage
- Backup
- Encryption keys
4 · Compute and the rest — last, once the risky foundations are reproducible
- Services
- Queues
- Load balancers
- DNS
Our role
What we automate and what we insist you keep
- 01
The drift report first
We start by showing what the console holds that no code describes, so the work is prioritised by risk rather than by whichever service was easiest to write first.
- 02
The pipeline, and the closed door
We build the plan-review-apply path and, with your security team, remove the standing console write access that lets drift back in. The second part is the one that makes the first worth doing.
- 03
A practice your team runs
We hand over modules your engineers can read and extend, not a black box only we can change, because infrastructure only you can operate is a dependency dressed as an asset.
The emergency-change policy and who may still touch the console stay your security team’s to set; we build the pipeline, the drift detection and the modules, and make the disciplined path the easy one.
Common questions
What buyers ask
We wrote Terraform once and it already does not match reality. What now?+
That is the usual starting point, not a failure. We run a drift report to measure the gap, then reconcile in the safe direction — importing what exists into state rather than letting an apply delete live resources — and bring the highest-risk parts back under code first. The lasting fix, though, is closing the path that caused the drift: while anyone can change infrastructure by hand, any reconciliation is temporary.
Does this mean nobody can ever touch the console again?+
For the accounts that matter, human write access through the console is removed, but a fast, recorded path for a genuine emergency stays — the aim is to make the disciplined route the easy one, not to block the 2am fix. Read access is untouched. The rule is only that a change leaves a trace and returns to code, because an emergency edit that never makes it back into the code is exactly how drift starts.
How does this help with data residency and compliance?+
When infrastructure is code, the residency, encryption and backup settings a regulator or a customer review asks about are written down and reviewable rather than remembered and hoped for. A reviewer can read exactly where a data store lives and how it is protected, and a change to any of it goes through a recorded plan. That turns a compliance question from an archaeology exercise across the console into reading the code and its history.
In this cluster
Next step
Map it to your estate
A drift report on your existing cloud — what the console holds that no code describes — and a plan to bring the highest-risk parts under code first.
