Solutions

Software across industries

We start with how your operation already works: the systems, the data, the workarounds. Then we deliver what is missing — software, the connectivity beneath it, or both. We work this way with finance, manufacturing, healthcare, retail, logistics and public-sector organisations alike.

Solutions

Our own products

We build software products that solve real problems in operations and keep working for years. Each product does one job properly, is supported by the team that built it, and can be bought and run on its own.

Middleware layer

Energy meters and machines, connected to your systems.

The middleware layer sits between the plant floor and everything above it. It reads meters and machines over the protocols they already speak, and publishes the readings in the formats your other systems can consume — so no two systems need to agree on a vendor’s idea of a file format.

What it reads

Modbus TCP / RTU

The common ground of meters, drives and small controllers.

OPC UA

Newer controllers, SCADA layers and standard industrial servers.

S7

Direct exchange with controllers that run S7.

PROFINET

Cyclic I/O and diagnostics on industrial Ethernet.

CC-Link

Field and industrial Ethernet networks.

EtherNet/IP

Industrial Ethernet environments that already use EtherNet/IP.

MQTT

Sensors and gateways that already publish over messaging.

Optocoupler I/O

Where a machine has no usable network port at all.

How the data leaves

REST / JSON

The standard shape for an application or a cloud service to consume.

Message queues

Streaming to a broker so several systems can subscribe independently.

OPC UA server

Present the data back as an industrial source other plant systems can read.

Database

Write straight into the operational database downstream.

Files

Periodic exports for reporting, reconciliation and archives.

Live streams

Push to dashboards that need the value as it changes.

Energy meters are read continuously and attributed to the line or machine that consumed the power — consumption per line, per shift, and per site, without anyone walking the plant with a clipboard. The readings are exported from the middleware layer in whichever shape your systems expect, so reporting, billing and maintenance all work from the same numbers.

Deal assessment

Price the deal on what is happening now — and on everything that happened before.

Large deals still get approved on experience and a spreadsheet. This turns that judgement into a score the desk, the risk function and the auditor can all read: the deal as it stands today, plus the record of how comparable deals actually ended.

What it weighs

Counterparty

Who is on the other side, and how much exposure already sits with them across open deals.

Live market

Current prices, spreads and volatility, read as they move rather than once a day.

Deal structure

Volume, tenor, delivery point, payment terms and the guarantees that sit behind them.

Historical outcomes

How comparable deals performed, read back from the archive of past transactions.

Limits & mandate

What the desk is authorised to take on, checked at the moment of the decision rather than after it.

Credit & collateral

Credit standing, margin and security, carried in the same score as every other factor.

What it produces

One risk score

A single number per deal, with the factors behind it visible and openable.

Approve, escalate, decline

Routed to the limit holder with the authority to sign, not to a queue.

An audit trail

Every assessment stored with the data that produced it, so a decision can be reconstructed years later.

Re-assessment on change

A deal that moves is scored again. Approval is a position, not a moment.

Two clocks, one system. New data arrives continuously; the historical record is kept in S3-compatible object storage and read back on every assessment. The score therefore weighs today's deal against a current picture and against everything the archive knows — not against a table refreshed last quarter.

LLM Generation Gateway

One controlled door to every language model you use.

Teams adopt language models faster than a company can govern them. The LLM Generation Gateway sits between your people and every model they call: prompts leave through it, answers come back through it, and the policy that decides what is allowed is written once, in one place.

What it guards

What leaves

Sensitive fields are detected and redacted before a prompt leaves the perimeter, whichever application sent it.

What comes back

Answers are checked against policy before they reach a person or feed a downstream system.

Who asked what

Every request, model, prompt and response is logged with the user and the stated purpose behind it.

What it controls

Which model serves which call

Routing is decided by policy rather than by whichever application happened to be written first.

Cost and volume

Spend is visible per team, per use case and per model, with limits that hold.

Regulated environments

Built for banks and companies that must show a supervisor what the model was asked and what it answered.

The gateway is the layer between your applications and the model providers. Every call goes through it: it authenticates the application, applies policy — redacting sensitive fields, allowing or blocking a model, enforcing rate and spend limits — forwards the request, checks the response against the same policy, and stores one record per call with the user, the purpose, the model, the prompt, the answer, the latency and the cost. Applications point at one endpoint and hold no provider keys, so a model swap, a blocklist change or an audit export is a central configuration change rather than a release.

What we bring

Three layers, one supplier.

We own all three layers. That is why a fault can reach a technician’s phone while it is still happening.

01Software & apps

Software

We write reliable softwares: business applications and the integrations between them. This is the layer that turns operational data into records, work queues, reports and alerts — on the web and on mobile.

  • Custom applications and integrations
  • Modernisation of existing systems
  • Web and mobile apps
  • Dashboards, alerts and reports
02Protocols

Middleware

The layer that speaks industrial protocol natively and normalises it, so machines from different vendors finally share one model of the plant.

  • PROFINET
  • Modbus TCP / RTU
  • CC-Link IE
  • S7, OPC UA, EtherNet/IP
03Shop floor

Hardware

Our own communication units, installed on the line, holding a real-time channel to the controllers already in your cabinets.

  • Any PLC, any make
  • PROFINET · Modbus · CC-Link IE
  • Optocoupler I/O where needed
  • Energy meters via the middleware layer
Where we work

Many fields, each with its own rules.

Every industry has its own rules. We learn the field before we propose anything — the constraints shape the architecture from the first diagram.

01

Manufacturing & Production

Where the signals are, and where they usually stop.

Production planning, machine integration, and quality systems that connect the shop floor to the ERP in real time — including the controllers and meters already installed.

  • Live machine states and error codes
  • Production and OEE tracking
  • CMMS with GANTT scheduling
  • Energy consumption per line
02

Logistics & Supply Chain

Moving things reliably, at scale.

Warehouse, fleet, and supplier orchestration with real-time visibility — stock levels that match what is physically on the shelf, not what the spreadsheet says.

  • WMS with integrated inventory
  • Shipment and route tracking
  • Supplier and procurement portals
  • Predictive stock planning
03

Finance

Systems where a rounding error is a legal problem.

Transaction platforms, ledgers, reconciliation engines, and regulatory reporting — now with operational cost fed in from the plant rather than allocated at month end.

  • Multi-currency ledgers and settlement
  • Audit trails with immutable logs
  • Regulator-ready data models
  • Cost per machine, per intervention
04

Asset & Maintenance

Everything the plant owns, and what it costs to keep.

Asset registers and maintenance systems covering machinery and equipment from commissioning to replacement, with downtime measured instead of estimated.

  • Full asset lifecycle
  • Preventive and reactive work orders
  • Spare parts and procurement
  • Downtime measurement
05

Healthcare & Life Sciences

Privacy-first systems for regulated environments.

Clinical and administrative platforms built around consent, traceability, and compliance from day one.

  • Patient and clinical records
  • Compliance (GDPR, HIPAA-aligned)
  • Interoperability with existing systems
  • Anonymised analytics
06

Public Sector

Software for institutions that cannot afford to fail.

Citizen-facing platforms, internal workflows, and data consolidation designed for reliability, accessibility, and long-term maintainability.

  • Digital service delivery
  • Inter-agency data exchange
  • Long-term maintainability
  • Accessibility and openness
How we deliver

Five stages, from first look to support.

It starts with a look at the systems you already run.

  1. 01

    Start where the work happens

    We start with the process itself, not in a meeting room: which systems are involved, what they cannot see, and where the work is still done twice by hand.

  2. 02

    Prove it on one case

    A small pilot on one process, one machine, or one reconciliation. If the data cannot be reached, we say so early — before you are committed to a project.

  3. 03

    Deploy one module

    Usually CMMS or energy. It goes live, it is used daily, and it starts producing data you did not have before.

  4. 04

    Integrate the rest

    Additional modules are added on the same platform, sharing the same data — finance, stock, HR, projects, compliance.

  5. 05

    Stay involved

    Monitoring, tuning, and evolution. We built the hardware the data comes from, so support has no hand-off.

How we build

How we build.

We choose our tools with the same discipline we ask of everything else: mature, well-documented, actively maintained, and — in the plant — actually present on your equipment.

Enterprise-grade foundations

Technology chosen for a decade-long support horizon, not for this year’s trend.

Protocol coverage

The plant protocols your equipment already speaks, handled directly.

Interfaces people can use

Operational screens built for eight-hour shifts, not demos.

Real-time rules and alerts

Events judged as they arrive, so alerts are useful rather than retrospective.

Models on your own data

AI tuned to your operation and run inside your perimeter.

Integration, not replacement

Built to sit alongside the systems you already run.

Tell us what nobody can see.

Describe the system, the data source or the process. We will tell you what is reachable and what it would take.