Rafiq · Pitch Deck
Defence & Security · OPERATOR AI
Rafiqرفيق
Deterministic AI copilots for triage, planning, and after-action analysis — inside your enclave.
neuriya.ai — Sovereign, Arabic-first enterprise AI
Problem
Analysts triage, plan, and report at human speed while the volume of intelligence keeps rising
- Incoming HUMINT, SIGINT, and IMINT reports pile up faster than analysts can read them, so urgent signals sit in the same queue as routine ones until someone gets to them.
- First-pass mission plans start from a blank page every time, even when the underlying intelligence picture has barely changed from the last planning cycle.
- After-action reports are written from memory and notes days after a mission, once analysts are free, so details fade before they're recorded.
- AI tools built for civilian use don't offer deterministic, reproducible outputs, so a recommendation can't be re-run and verified the same way twice during audit.
Solution
Deterministic AI copilots for triage, planning, and after-action analysis — inside your enclave
Product
Rafiq in action
How it works
From first touch to outcome
Ingest
Reports and mission data flow in from existing systems without any change to source-side collection.
Triage
The copilot pre-sorts by urgency and source corroboration, surfacing what needs eyes first.
Draft
A first-pass mission plan or after-action report is generated for the analyst to shape, not to rubber-stamp.
Approve
Nothing moves forward without a logged analyst sign-off — the copilot recommends, the operator decides.
Market
Sizing the opportunity
- TAM — $2.6B. GCC defence AI decision-support & analyst-tooling market
- SAM — $300M. Saudi military and security AI copilot & analyst-augmentation spend
- SOM — $20M. 3-year obtainable share via design-partner unit deployments
Vision 2030's military modernization program is pushing units to adopt AI-enabled operations, while doctrine and NCA ECC-2 requirements keep human sign-off mandatory — a combination that favors a deterministic, enclave-native copilot over a generic assistant retrofitted for defence use.
Business model
How Rafiq makes money
Per-analyst-seat license
Priced per active analyst or operator seat using the copilot inside the accredited enclave.
Enclave deployment subscription
Recurring fee covering the on-prem model and inference stack running inside the unit's accredited enclave.
Implementation & enclave integration fee
One-time fee to integrate the copilot with existing intel, planning, and reporting systems.
Revenue streams
Where the revenue actually comes from
Competition
Where we win
Our edge
- Deterministic, reproducible outputs are the design starting point, not a constraint added after a generative-AI prototype.
- Runs entirely inside the accredited, air-gap-capable enclave — no data or query ever leaves it, unlike copilots built on external model APIs.
- Human sign-off is structural: no recommendation reaches a plan or report without a logged analyst decision against their identity.
- Founding team combines Saudi defence and intelligence-community operating experience with applied AI engineering.
Traction
Early validation, built for pilot readiness
- Full mission-copilot experience designed and built end-to-end: active-missions dashboard, intel triage board, and case file with sign-off trail.
- Architecture pre-aligned to NDMO Classification, NCA ECC-2, ISO 27001, and an accredited enclave deployment model.
- Actively selecting design-partner units for a first accredited pilot cohort.
Go-to-market
Design-partner unit first, then enclave-wide analyst rollout
- 1Land with 1-2 design-partner units willing to deploy the copilot inside an existing accredited enclave for a defined analyst group.
- 2Prove triage-speed and mission-planning quality gains over a pilot period, with every recommendation traceable to a logged analyst sign-off.
- 3Expand to additional analyst seats and mission types within the same unit, then to further units through the accreditation path the first pilot establishes.
Team
Built by operators, not just builders
The ask
Raising a seed round to reach first accredited unit pilots
We're raising to fund enclave-deployment tooling, deterministic-model hardening, and the core engineering team through first-revenue pilot deployments.