Case Study: Argus
for Genix Cyber
Introduction
MetroMax built Argus for Genix Cyber end to end, then designed AI into it: real-time, context-aware risk scoring and an analyst assistant grounded in the platform’s own data.
At a glance
The challenge
Genix Cyber needed a security platform built around its own product, not a feature added to an existing system.
Surfacing alerts was not enough. Security analysts face far more alerts than they can act on. The scarce, expensive skill is deciding which alerts matter and what to do about them. Argus had to help with that decision, not just add to the queue.
What we built
Argus is three layers, each built by MetroMax.
The platform. Built end to end for Genix Cyber: architecture, data model, event pipeline and interface.
Real-time risk prediction with context. Events are scored as they arrive, using the surrounding situation rather than isolated signals. Prioritisation reflects what is actually happening, so analysts see the alerts that matter first.
AI assistant for analysts. A chat interface for interrogating alerts: what happened, what it means and what to do next. Answers are grounded in the platform's own data, not general knowledge.
Our approach
We built the platform first, then built the AI into it. Because we owned the data model and event pipeline, the risk model and assistant could draw on the full context of each event from day one.
Designed in, not bolted on. The AI layer uses the platform's own data structures rather than a separate integration.
Grounded answers. The assistant responds from Genix Cyber's platform data, which keeps its guidance specific and inspectable.
Analysts stay in charge. The AI prioritises and explains; the analyst decides what action to take.
This gave MetroMax first-hand knowledge of what it costs to design AI into a product versus retrofitting it later.
Outcomes
Genix Cyber has a purpose-built security platform with AI prioritisation and investigation built into the analyst workflow, rather than added afterwards.
Measured results are not yet in this draft. Figures to confirm with Genix Cyber before publishing:
Reduction in alerts needing manual review
Change in mean time to triage or respond
Analyst adoption of the AI assistant (queries per analyst per week)
Analyst adoption of the AI assistant (queries per analyst per week)
Why it transfers
Alert triage, and the request to “explain this to me and tell me what to do”, has the same shape as a compliance analytics assistant. The same pattern applies wherever users face more signals than they can act on, such as fleet compliance and driver risk.
MetroMax is a product engineering firm that builds AI into things, not an AI vendor looking for something to bolt onto.
