01

Cloud and platform

Where systems run, how environments are separated, and how any of it is reproduced.

Microsoft Azure
Subscription and resource topology, networking, policy and cost structure.
Infrastructure as code
Bicep and Terraform. Environments are defined, reviewed and reproducible.
Networking
Connectivity between cloud, offices and existing on-premise systems.
Containers and compute
Chosen against the operational burden a client can actually carry.
Other clouds
Where an organization already runs on AWS or Google Cloud, we work there rather than propose a migration for its own sake.
02

Identity and access

The layer everything else authenticates against, and the one most estates get wrong first.

Microsoft Entra ID
Directory architecture, conditional access, privileged access and review.
Single sign-on
SAML and OIDC across internal and third-party applications, including ones never designed for it.
Lifecycle and provisioning
Joiner, mover and leaver as an automated process rather than a checklist.
Federation
Access to systems outside the organization’s own directory.
03

Workplace platforms

The platforms an organization already runs on, treated as managed infrastructure.

Microsoft 365
Tenant configuration, governance and lifecycle across Exchange, SharePoint and Teams.
Information governance
Retention, classification and legal hold, applied where records actually live.
Device management
Intune and endpoint policy, including contractor and shared-device cases.
Business process automation
Approvals and provisioning automated inside the platforms already in use.
04

Applications

The software an organization puts in front of people.

TypeScript and React
Next.js for web platforms and customer-facing products.
.NET and Node
Server-side services, chosen to match what a client can maintain alongside us.
Mobile
Native and cross-platform, with release and distribution operated alongside the build.
API design
Interfaces treated as products: versioned, documented and monitored.
05

Data and integration

How systems agree with each other, and how anyone finds out what happened.

Relational data
SQL Server and PostgreSQL, with migrations under the same review as application code.
Integration patterns
Queues, events and scheduled transfer, selected against real failure modes rather than fashion.
Reporting layers
A defensible position on where a number came from, before anyone builds a dashboard on it.
Legacy interfaces
EDI, file transfer and vendor APIs, because most estates still depend on them.
06

Operations tooling

What makes the difference between a system that runs and one that is operated.

Pipelines
GitHub Actions and Azure DevOps. Releases are repeatable and reversible.
Observability
Knowing something is wrong before the business reports it, with enough signal to know why.
Backup and recovery
A recovery position that has been tested, not merely configured.
Secrets and configuration
Held in managed stores in the client’s own tenancy.
Stated plainly

No badges we have not earned.

Ferrafox holds no vendor partner status and no platform certifications, and displays none. Listing a platform here means we build on it and operate it, not that its vendor endorses us.

Where an organization already runs on something outside this list, that is usually the right place to keep it. Recommending a migration because it suits the supplier rather than the estate is how the last decade of technical debt got there.

09Contact

Tell us what your organization runs on.

The first conversation is about the environment as it stands today: what exists, what it costs to keep running, and where responsibility currently sits.