Security consultancy · founder-led

Security built in layers, around the way you already work.

For companies that need to pass an audit, clear an enterprise security review, harden their cloud or detect threats, without building a full in-house team. Working with clients in the US, East Africa and beyond.

Book a scoping call30 minutes. You leave with a written scope.
DEFINEGovernance, risk and complianceBUILDSecurity engineeringPROVESecurity operations
Fig. 1. An alert, followed to the endILLUSTRATIVE · EXAMPLE CLIENT OUTPUT
ALERT · 09:41 ET
Break-glass sign-in from a new network
LAYER 1 · AGENT
Context gathered, signals correlated
LAYER 2 · ANALYST
Containment approved by a person
LAYER 3 · CONTROL
Gap closed in engineering
LAYER 4 · EVIDENCE
Proof filed for the auditor
INVESTIGATION · ALRT-0412OPEN · DETECTED
09:41 DETECTIONConsole sign-in to production with the break-glass role, from a network not seen in 90 days.
09:43 AGENTPending
09:52 ANALYSTPending
11:20 CONTROLPending
11:34 EVIDENCEPending
Agents investigate. People decide. Every containment action above was approved by a person.

What we do

Three layers, one system.

Governance defines the controls. Engineering builds them into the systems you run. Operations shows they hold. Engage one layer on its own, or all three.

A crack that gets past one layer meets the next, and turns aside. What operations learns flows back into the controls.

LAYER 1 · GOVERNANCELAYER 2 · ENGINEERINGLAYER 3 · OPERATIONS
TURNED ASIDE · LAYER 1
Production access needs an approved request
TURNED ASIDE · LAYER 2
Guardrail blocks the public storage bucket
TURNED ASIDE · LAYER 3
Detection fires, agent writes the summary
STOPPED
Analyst approves containment. The fix flows back to layer 1.
Fig. 2. A crack turned aside, illustrative
01Define

Governance, risk and compliance

Risk assessments, policies and compliance programmes for the frameworks your customers and regulators ask for, such as SOC 2, ISO 27001 or NIST CSF. The evidence comes from the controls themselves, so audit season stops being a spreadsheet hunt.

YOU HOLDRisk register, policies your team can follow, a control map per framework, evidence from live systems

02Build

Security engineering

Cloud security, identity and access, logging and telemetry, infrastructure-as-code guardrails and security automation. Built into the systems you already run, with your tools or ones we recommend.

YOU HOLDA baseline report per account, guardrails merged as code, a logging coverage map, runbooks for your engineers

03Prove

Security operations

Detection engineering, response playbooks and agentic SecOps. AI agents gather context and write the summary; a person approves every containment action. We build it and hand it over, run it with your team, or operate alongside you in stated hours.

YOU HOLDDetections with test cases, response playbooks, investigation records, a monthly evidence pack

EMERGING

AI security and governance. Guardrails and evaluation for the models and agents you build, so they act within the limits you set. A growing part of our work.

FULL-BLEED MACRO · NACRE CROSS-SECTION · OWN OR LICENSED
Mineral tablet · about 0.5 µm thick
Plate 3. The name

A crack cannot run straight through nacre. Each soft layer turns it aside until it stops.

Nacrion, said NAY-kree-on, is named after nacre, the material that lines a shell. We build security programmes on the same principle. No single layer has to be perfect, because the structure holds.

We layer around your shape.

Nacre does not choose the shape of its shell. It forms against whatever surface it lines, and it is strongest where it fits most closely.

We start from your organisation, your systems and the tools you already run, then add layers where they are needed. Scope, pace and the frameworks you answer to all follow you. Some things never move.

Fig. 4. Start from what you need to doSomething else? Tell us your shape.
YOUR ORGANISATION
As it is today
GOVERNANCE
Policies and control map
ENGINEERING
Controls the auditor will test
OPERATIONS
Monitoring evidence
ROOM TO
OPERATE
GovernanceEngineeringOperationsILLUSTRATIVE
A LIKELY STARTING POINT

Audit readiness programme

Ready for a SOC 2 or ISO 27001 audit, with evidence drawn from controls that run every day.

LAYERS
Define + Build
TYPICAL
8 to 12 weeks
TOOLS
Yours, or ones we recommend

Adjustable. We shape scope and pace with you on the call.

Book a scoping call →
FIXED BEFORE WE START
  • The deliverables, listed by name
  • The outcome we are working toward
  • The evidence you hold at the end
  • The cost, agreed in writing
SHAPED TO YOU
  • Scope, pace and the order of work
  • The frameworks you answer to
  • Your tools, or ones we recommend
  • How closely your team is involved
Fig. 5. Ways to work. Select a position.CLIENTS MOVE RIGHT AS THEY GROW

We design and build alongside your engineers.

WE DO
Architecture, pairing, code review
YOUR TEAM
Builds with us and owns the result

Starting points

Every engagement begins from one of these and is shaped in a scoping call. Names, scope and timelines are typical and all of them adjust.

A

Cloud and identity baseline

BUILD

Your cloud accounts and identity provider meet a written baseline, with drift alerts in place.

Account structure, IAM and SSO review, logging, guardrails as code.

4 to 6 weeks
typical · adjustable

B

Audit readiness programme

DEFINE + BUILD

Ready for a SOC 2 or ISO 27001 audit, with evidence drawn from controls that run every day.

Gap assessment, policies, control mapping, evidence collection, auditor liaison.

8 to 12 weeks
typical · adjustable

C

Detection and response build

PROVE

Detections for your highest risks, written playbooks and agent-assisted triage your team can run.

Telemetry review, detection engineering, playbooks, handover or shared running.

6 to 10 weeks
typical · adjustable

D

Layered partnership

DEFINE + BUILD + PROVE

A standing programme across all three layers, paced to your roadmap.

Quarterly plan, monthly evidence pack, a named engineer.

Rolling
reviewed quarterly

Something else? Tell us your shape.Book a scoping call →

What we hold ourselves to, and how you can check.

Layered

No control is trusted alone.

YOU CAN CHECKYour control map names the second control behind every critical one.

Calm

Measured language and plain reporting.

YOU CAN CHECKReports give severity in words and numbers. Nothing depends on a red light.

Evident

Every claim we make can be proven.

YOU CAN CHECKEach deliverable ships with its configuration exports, logs, test results and sign-offs.

Resilient

Stronger after an incident, not just restored.

YOU CAN CHECKEvery incident closes with a written record of the change we made and the test behind it.

Who you work with

Founder-led. The person who scopes your work is the person who builds it.

Nacrion is led by an engineer who has worked in every layer of a security programme. That means writing the policy and mapping the controls, building the cloud and identity guardrails, then running the detections that show they hold. One person can follow a gap from the audit finding to the fix to the evidence.

GOVERN
Risk, policy and compliance programmes
ENGINEER
Cloud, identity, logging, infrastructure as code, automation
OPERATE
Detection, response and agent-assisted investigation
EMERGING
Securing AI platforms and agentic systems
CLIENTS
Global. In the US and East Africa today, and more regions as we grow.

We are new, so we show our work instead of logos.

Before you sign, you can read our written method, see a redacted sample deliverable and get a fixed scope and cost in writing.

Ask for the sample deliverable
CONTROL COVERAGE · EXAMPLE CLIENT OUTPUTILLUSTRATIVE
SOC 2 · Security criteria41 / 46 live
ISO 27001 · Annex A72 / 93 live
Cloud baseline checks118 / 124 pass
Detections with passing tests14 / 18 tested
Numbers belong to a fictional client and show the format only. "Live" means evidence collected from the running control within the last 30 days.

Tell us about your security shape.

A 30-minute scoping call, then a written scope.

Book a scoping call