Skip to content
OXWYN Contact

Oxwyn · Intelligence and R&D

From requirement to fielded capability.

We take a mission requirement and turn it into a fielded capability for defence, security, critical infrastructure and health. Everything we build stays on site, leaves the final decision with a named person, and records every decision.

Mode · Human in command
Data · Stays on site
Record · Tamper-evident
Site · Full website soon

The method

We build what the mission needs, not what a catalogue offers.

Every Oxwyn capability starts with the people who will use it and the rules they work under. Six phases take a requirement from the first question to support in the field.

  1. PHASE 01

    Requirement

    What decision must be made, by whom, under what pressure, and what failure would cost.

  2. PHASE 02

    Doctrine

    How the user actually operates: authority, chain of command and procedures.

  3. PHASE 03

    Tactics

    The conditions on the ground, the threats faced and the time available.

  4. PHASE 04

    Research

    Intelligence, analysis and design: architecture, security boundaries and the human role.

  5. PHASE 05

    Build

    Software and purpose-built AI models, trained on the customer’s own data and run on site.

  6. PHASE 06

    Assure

    Independent testing, evidence for accreditation and support in the field.

Our doctrine

Four rules in everything we build.

RULE 01

It stays on site.

Sensitive data does not leave the building, base or hospital.

RULE 02

A person decides.

A named, authorised person owns every consequential decision. The system recommends; it never signs.

RULE 03

Every decision is recorded.

A tamper-evident record of what was decided, by whom, and on what evidence.

RULE 04

Tested where it matters.

Against rare, high-consequence events, not only the averages. We state our status honestly.

Where we work

  • ADefence
  • BPublic safety and security
  • CData centres and critical infrastructure
  • DHealth

Purpose-built AI

We build language and AI models for defence, government and health, trained on the customer's own data and run under the customer's own control.

Aligned to the standards each customer requires, including ISO/IEC 27001 and 42001, the NIST AI Risk Management Framework, NIST SP 800-171, the EU AI Act and Secure by Design.

Contact

Have a requirement that off-the-shelf technology cannot meet?

We work with defence and government teams, integrators and research partners. Tell us about your site and the decision you need to trust.

hello@oxwyn.com