Software building & security development · UK

Software that does the work while you sleep.

Monasterial designs and builds bespoke AI systems, platforms and mobile apps, from job allocation engines and automated sales outreach to neural-network intrusion detection. We start with a costly problem in your business and build the system that handles it every day.

Systems runningWhat the systems we’ve built automate
Founder

Hi, I’m Hamza.

I do software building and security development. I founded Monasterial to build systems that take real work off people’s plates, and I’m obsessed with making projects.

I’ve just completed a cyber security scholarship, which is why security runs through everything I ship: private storage, access control, encrypted channels and systems that are tested against real attacks. Alongside client platforms, I build my own research projects, like Project Strike, an intrusion detection system driven by a spiking neural network.

Next I’m aiming to start a PhD in spiking neural networks within the next year. Pulse and Watchtower are part of that path.

BackgroundCyber security scholarship, completed
What I doSoftware building & security development
ResearchSpiking neural networks · PhD next
BasedUnited Kingdom
Security & AI research

Networks that
watch the wire.

Security runs through everything we build. This is our in-house security and machine learning work, starting with an intrusion detection system built around a spiking neural network.

SECURITY 01

Project Strike

Built by MonasterialNetwork intrusion detection · Spiking neural network

A hybrid network intrusion detection system designed and built from the ground up. A spiking neural network reads the rhythm of the traffic, a rules engine checks hard signatures in parallel, and every alert leaves the sensor through an encrypted, tamper-evident channel to a SIEM.

Neuromorphic detection

Live traffic becomes a 13-feature behavioural vector covering protocol rates, port entropy, sequential-port ratio, burstiness, inter-arrival timing and TCP flag anomalies. It is rate-encoded into spike trains over 25 time-steps and fed through three layers of leaky integrate-and-fire neurons (64 → 32 → 5), trained with surrogate gradients in snnTorch.

Hybrid rules + fusion

A YAML signature engine runs alongside the network. When both engines flag the same attack within a two-second window, the alerts are fused into one full-confidence alert and duplicates are suppressed.

Campaign correlation

Alerts are linked over five-minute windows into multi-stage campaigns: reconnaissance cascades (ARP scan → port scan), scan-then-flood sequences, and many sources converging on one target.

Encrypted alert channel to SIEM

A 2048-bit Diffie-Hellman key exchange with SHA-256 key derivation sets up the channel. Alerts are sent with AES-256-CBC and HMAC-SHA256 (encrypt-then-MAC), with a 60-second replay window. The SIEM only accepts alerts that pass the integrity check.

Adversarial test range

A Dockerised station and attacker network runs real ARP scans, UDP port scans, TCP Christmas-tree probes and UDP floods to prove detection end to end.

PythonPyTorchsnnTorchScapyPyCryptodomescikit-learnDocker
OutcomeIn live attack tests it caught UDP floods, stealthy ARP reconnaissance and UDP port scans. Every alert reached the SIEM encrypted and integrity-checked.
strike / station 172.20.0.2Monitoring
Launch traffic at the sensorauto-play
CaptureFeaturesRulesSNNFusionEncryptSIEM
Capture
Pick an attack above to watch it go through every engine.
Output neurons · membrane potentialdashed line = threshold
In the lab

Where the Strike engine goes next. Both demos below are live, so try them.

Active R&DSECURITY 02

Pulse

Spiking-neural anomaly detection for AI products.

Pulse points the Strike network at product usage instead of network traffic. A leaky integrate-and-fire neuron builds up charge from each account’s request pattern and fires on behaviour that fixed limits miss.

account #2291 · requests / minnormal usage
Fixed rate limitNo alertTraffic stays under 120 req/min
Pulse neuronQuietMembrane below threshold
snnTorchTime-series featuresSupabaseCost metering
Active R&DSECURITY 03

Watchtower

A SOC and SIEM lab for detection engineering.

Watchtower replays real adversary techniques against a segmented Windows and Linux lab. Each technique is mapped to MITRE ATT&CK and detected by Wazuh, Suricata, Sigma rules or Project Strike. Any gaps are closed with new rules.

att&ck coverage · lab-01coverage: not run
Run a chain, or click any technique.
WazuhSuricataAtomic Red TeamMITRE ATT&CKSigma rules
Selected client work

Three businesses.
Three bespoke AI systems.

AI is built into every platform we ship. Each of these systems was designed and built by Monasterial around one specific commercial problem, with a live look at how each one works.

CASE 01 / 03

Inspired by the Caribbean

Built by MonasterialLuxury property · Barbadosinspiredbythecaribbean.com

An intelligent property search platform with its own CRM and booking system. Each visit adds to a picture of what the customer wants, and the platform keeps re-engaging them with properties that fit it.

Behavioural buyer profiling

Saves and enquiries count as signals of intent. The system finds what a customer’s saved villas have in common, from bedrooms and bathrooms to private pools, and refines the profile as they browse.

Automated weekly matching

Every week each profile is searched again, and customers get a personal email of newly matched properties with no work from the sales team.

Intent-triggered outreach

When interest spikes on a property, the AI starts a conversation by WhatsApp, phone or email while the buyer is most receptive.

Language & dialect adaptation

It picks up the customer’s language and dialect and replies in kind, which matters for an international audience making a high-value purchase.

OutcomeProfiles, conversations and reservations live in one bespoke system, and the business stays in touch through a long buying cycle without extra staff.
profile_engine / customer #2291Learning
1 · Saved by the customer0 saved
West coast villa5 bed · 5 bath · pool
Beachfront house4 bed · 4 bath · pool
Hillside estate6 bed · 6 bath · pool
2 · Preference profileconfidence
Bedrooms 4+0%
Private pool0%
West coast0%
Sea view0%
3 · Re-engagementautomatic
Weekly email · 3 new matches
5 bed · pool · sea view98%
4 bed · pool · west93%
5 bed · pool · garden88%
WhatsApp · high intent detected
Hola, we saw you’ve been looking at the west coast villa again. It has the private pool and five en-suites you wanted. Would you like a video tour this week?
CASE 02 / 03

ReadyClean

Built by MonasterialCleaning services · United Kingdomreadyclean.co.uk

A platform that runs the three jobs that eat the most admin time in a field-services business: recruitment, customer bookings and payroll. The system does the processing and people provide oversight.

Intelligent job allocation

When a booking lands, an AI scheduling engine checks every cleaner’s calendar and how far each one is from the client, then assigns the job.

Points-based decision model

A bespoke scoring model weighs availability, proximity and scheduling efficiency to pick the best match. That cuts travel and fuel, fits more jobs into each day and takes the guesswork out of the rota.

Recruitment & payroll automation

Onboarding, assigned work and pay calculations all run on the same platform, so admin doesn’t grow as the team does.

OutcomeRunning reliably since launch. Less travel between jobs, more jobs per cleaner per day, and a rota that plans itself.
allocation_engine / booking #4821Scoring
New job · 2h clean
CleanerAvailableProximityEfficiencyPoints
CASE 03 / 03

Cachi AI

Built by MonasterialSEO automation · Australiacachiai.com.au

An SEO platform for Australian trade and blue-collar businesses. It keeps reviewing, updating and improving each client’s search optimisation and website content, as an ongoing service rather than a one-off job.

Continuous, automated optimisation

The system checks and refines content and SEO performance at regular intervals, so visibility holds as search conditions and competitors change, without the client commissioning new work.

Built around the market it serves

Tradespeople rarely have time for technical or digital work and most would rather outsource it completely. Cachi AI gives them strong search visibility with no technical involvement at any stage.

OutcomeSearch visibility that keeps improving after launch. Clients get the result without logging into anything.
seo_agent / optimisation loopOptimising
RunsEvery week
CoversWhole site
Client effortNone
Search visibility trendcontinuous
HighLowWk 1Wk 6Wk 12
Week 1Technical audit · crawl issues fixeddone
Week 3Service pages rewritten for local searchdone
Week 6Added FAQ & local-business schemadone
Week 9New suburb pages from search demanddone
Week 12Refreshed content that competitors overtookdone
Products

Four apps.
One proven pipeline.

Each app puts the same production pipeline into a different industry: a user uploads something, it’s stored securely, AI processes it, a structured report comes back, and usage is tracked against plan limits. Pick an app and click through its screens.

Designed in-house by Monasterial
PRODUCT 01

AI Form Coach

How it works
How we work

From a costly problem
to a system that runs.

Every build starts with one commercial question: which work is costing you the most time or money? The AI is then designed to answer it.

1

Map the problem

We sit with how the work is done today and find where the hours, travel, missed leads or admin costs pile up.

2

Design the system

We decide the data, the decision logic and where people stay in control, then prototype it on real examples.

3

Build & harden

We build the platform or app with secure accounts, row-level security, private storage and tested access controls.

4

Run & measure

After launch we monitor jobs, costs and results, and keep tuning the system as the business grows.

Capabilities

The whole stack, under one roof.

Design, mobile, backend, AI and security from one studio. We’re security-minded because of where we come from.

Mobile apps

React Native and Expo, for iOS and Android from one codebase, with camera, uploads and a polished native feel.

AI & machine learning

Structured AI reports from video, audio and documents, plus custom models including spiking neural networks.

Platforms & CRM

Bespoke CRMs, booking systems, rota engines and payroll, built around how your business works.

Automation & outreach

Profiling, matching and outreach by email, WhatsApp and phone, triggered by what customers actually do.

Subscriptions

RevenueCat plans and trials, with entitlements and usage limits enforced on the backend.

Security & detection

Intrusion detection, SIEM pipelines, encrypted channels, row-level security, audit logs and access-control testing.

Start a project

Let’s put AI
to work in your
business.

Tell us which work costs you the most time. We’ll book a call to show how a system like the ones above could take it on, and what it would save.

© 2026 MonasterialAI systems · Platforms · Mobile appsinfo@monasterial.co.uk · monasterial.co.uk