For investors
Pre-seed. A full-lifecycle platform for building security systems — a large offline market that software has not truly reached yet.
The market: money everywhere, software nowhere
Security systems sit in every non-residential building: fire alarm and evacuation are mandated by law, video surveillance and access control are the de-facto standard. It is a durable market of design, hardware, installation and maintenance with recurring demand built in: systems age, codes get updated, buildings get commissioned and renovated.
Software in this market is stuck at the level of “a guard's console and a ticket log.” PSIM platforms aggregate alarms for large corporations, VMS manages video, CMMS tracks maintenance tickets — each tool covers its fragment, costs a fortune and does not talk to its neighbours. The mid-market — thousands of office buildings, warehouses, retail and clinics — is covered by almost nothing beyond spreadsheets and the contractor's word.
The fork in the road appeared recently: LLMs made the regulatory base — the hundreds of documents the whole industry rests on — programmable for the first time. Whoever first turns the codes into machine-readable rules with source citations, proven on real sites, gains an asset that followers will need years to rebuild.
Why now
Three waves are converging. Regulatory: digital as-built documentation and false-alarm accounting are turning from wishes into requirements — clients become obliged to have exactly the data the platform produces. Technological: LLMs have learned to work with regulatory text while preserving the reference to the source — something classic ML never managed. Cryptographic: the post-quantum transition is entering regulatory deadlines, and almost nobody in physical security is thinking about it — we arrive there first.
Meanwhile the barrier for competitors grows with every month of work: a machine-readable rule base + pilot-site data + equipment failure statistics is a compounding asset, not a feature that can be copied in one release.
The product through a revenue lens
M1 · Audit — revenue from phase one
A one-off service with a clear price tag: sellable right after the MVP, with no need to finish the rest of the platform. Every audit sold is also an entry ticket into the site for the subscription modules.
M3 · Monitoring — the subscription core
Recurring revenue per data point. Churn is low by the nature of the product: switching a monitoring platform is a project with installers on site, not a one-click cancellation.
M2 + M4 · Design and quantum — expanding the ticket
The design module monetises new buildings and renovations within the same client base. The quantum module sells into critical infrastructure and finance, where post-quantum readiness will be a requirement, not an option — and has no competitors in this bundle.
Monetisation model
The unit of pricing is the data point: a camera, a door, a detector. The client understands it without explanation (they know how many doors they have), it scales with the size of the site, and it grows as the platform covers more of the building's subsystems.
On top of the subscription — two streams: one-off audits and design work (cash from phase one, before the subscription economics kick in) and white-label for service companies. The platform hands them independent SLA metrics as a tender-winning argument, and hands us a sales channel into thousands of small sites without touching each one directly.
The network effect is baked into the data: the more sites under monitoring, the more accurate the predictive failure models — the more valuable the subscription for every next client. Failure statistics across a fleet of heterogeneous sites is an asset no hardware vendor owns.
In short
Stage
Pre-seed. A coherent methodology, written and estimated product requirements, a working portal skeleton. Raising for the MVP phase while lining up anchor pilots in parallel.
Business model
Subscription per data point + one-off audits and projects + white-label for service companies. First revenue in phase 1 (audits), subscription revenue from phase 2 (monitoring).
Moat
A machine-readable regulatory base with source citations, physical and IT security converged into one event stream, the post-quantum module, and a data network effect in predictive maintenance.
Horizon
MVP in ~3.5–4 months with a team of 5–6; the full platform in 4 phases, ~24 months, peak team of 10–11. Proving data-point unit economics is the phase-3 goal.
Risks — honestly
A conservative market
The industry does not buy from startups easily. Our answer: we enter through pain everyone acknowledges — false alarms, compliance orders, independent audits before real-estate deals — and through service companies, for whom white-label is a tender advantage.
The regulatory base keeps changing
For us this is not a risk but part of the business model: rules are versioned like code, base updates are included in the subscription, and every code change is a reason for the client to order a re-audit.
AI errors in a high-responsibility domain
Closed by architecture, not by promises: an answer without a citation of the code clause fails validation, and significant conclusions are confirmed by a human expert whose sign-off is recorded.
Financial model, effort estimates per phase, data-point unit economics and current status — on request.
Request details in the chatContact
Project news lives in the Telegram channel; questions and discussion — in the chat. We answer personally: investors, pilot clients and team candidates alike.