The portfolio

Products in active development.

Two agentic systems, both operating in domains where correctness is enforced rather than hoped for. Each is built on the same engineering discipline: narrow specialised agents, deterministic safety layers, grounded retrieval, and instrumentation that makes outcomes measurable.

Shared architecture

Different domains. The same engineering posture.

AVOLA and Forsight look nothing alike from the outside. Underneath, they are the same class of system, built against the same four constraints.

01

Narrow agents

Each agent has one testable responsibility, not a general-purpose prompt. Failures localise.

02

A safety layer

A deterministic check between the model's output and the user, enforcing domain rules the model cannot be trusted to remember.

03

Grounding

Recommendations trace back to validated protocols or verified organisational data, not to model priors.

04

Measurement

Longitudinal tracking that proves whether the system moved the number it claimed it would move.

Working on something adjacent?

We work with clinics, employers, HRMS platforms, L&D companies and staffing firms. If one of these products maps to a problem on your desk, we would like to hear about it.