Innovation

Research we run
alongside the software we build.

Most of our work is software we are asked to build. A smaller part is research we choose to do, because a problem is worth solving even when nobody has commissioned it yet.

Flagship initiative

InteliVis

An assistive technology initiative for people with visual impairments.

InteliVis is our attempt at a practical answer to a specific problem. We are exploring whether assistive technology for visual impairment can be made more accessible through ordinary cameras, phones and lower-cost hardware.

The direction we are exploring is a system that turns an ordinary camera and a phone or low-cost board into something that describes the user's surroundings, reads text aloud, and stays inside a budget a person or a clinic can actually afford.

We are publishing this because the work is real, not because it is finished. See AI & Innovation for how we scope research, what we deliberately do not do, and where the boundaries are.

InteliVis is an in-development research initiative. It is not a product, it is not available to buy, and we are not publishing a launch date we cannot stand behind.

A person wearing dark glasses holds up a phone on a city street, and the phone screen shows a pharmacy identified ten metres ahead. Alongside it, a diagram shows a camera capturing the world, the InteliVis system interpreting the scene, and the two results: reading text and spoken audio guidance.

InteliVis concept illustration. A design direction, not a working product.

What it is
An assistive system for people with visual impairments, designed around cameras and low-cost hardware rather than specialist devices.
What it is not
Not a shipping product, not a subscription you can sign up for, and not a claim of medical effectiveness.
Why we do it
Software. Security. Intelligence. This is the intelligence part applied to a problem our engineering discipline happens to suit.
Research & development

What else is in progress

Applied AI
Where artificial intelligence genuinely reduces work for the systems we build, rather than being added because it is available. AI & Innovation
Security research
Practical attack-surface and application-security work, published where it is useful to the people running systems. Insights
Published research
Peer-reviewed and archived work, including our IoT attack-vector paper. Read the research
Working with us

If you have a problem worth researching

We take on a small number of research-shaped engagements, particularly where the outcome matters more than the paperwork.