Digital product engineering · IoT

IoT platform development

We develop IoT platforms for companies that need to manage fleets of connected devices and use their data: a device registry and secure provisioning, connectivity over MQTT and other protocols, telemetry storage, device twins, rules and alerts, orchestration of over-the-air updates and APIs into your enterprise systems. We build the software and integration around your devices, and you own the code.
Drawing for IoT platform development
Checklist · IoT and devices

Industrial IoT and device integration checklist

A checklist for projects that connect devices, machines and sensors to software: use case, device inventory, protocols, gateways, data model, security, lifecycle and enterprise integration.
  • 01Eight areas to settle before you connect a fleet of devices or a plant.
  • 02The device and signal facts that integration teams always need.
  • 03Security and update items that are hard to add later.
  • 04Seven questions to put to any vendor or development partner.
Free download

Get it on screen in a few seconds.

No newsletter. We use your details to follow up once about the download, and you can ask us to delete them at any time.
What it covers

What the system does, module by module.

  • 01

    Device registry

    Every device with its identity, type, firmware version, owner, location, tags and lifecycle state, searchable and linked to customers and assets.
  • 02

    Secure provisioning

    Onboarding of devices with certificates or pre-shared keys, bulk and zero-touch enrolment, and revocation when a device is lost or retired.
  • 03

    Connectivity and protocols

    Brokers and gateways for MQTT, HTTP, CoAP and LwM2M, with authentication per device, topic-level permissions and support for constrained and intermittent links.
  • 04

    Telemetry ingestion and storage

    Validation and decoding of messages, time-series storage at the right resolution and retention, and replay of data after an outage.
  • 05

    Device twins

    A synchronised representation of each device's reported and desired state, so applications can read and set configuration whether or not the device is online.
  • 06

    Rules, alerts and automation

    Conditions on streams and states that raise alerts, open tickets or trigger actions in other systems, with throttling and escalation.
  • 07

    Commands and configuration

    Remote commands and configuration changes with delivery tracking, timeouts, approval for risky actions and an audit record.
  • 08

    Over-the-air updates

    Orchestration of firmware and software updates in staged rollouts, with health checks, automatic pause and rollback, using the device's own update mechanism.
  • 09

    Applications, APIs and multi-tenancy

    Dashboards, public APIs and webhooks, and tenant separation for platforms serving several customers or business units.
Architecture

Reference architecture.

IoT platform: reference architectureDevices and gateways connect securely through a connectivity layer to an IoT platform with a device registry, provisioning, telemetry ingestion, device twins, rules and alerts, commands and update orchestration. Applications, enterprise systems and analytics consume the data through APIs.DevicesIoT platformConsumersSensors and machinesTelemetry, eventsGateways and edgeProtocol translationMobile and field toolsProvisioning, serviceIoT platformDevice registryIdentity, state, tagsProvisioningKeys, certificatesTelemetry storeTime series, replayDevice twinsReported and desiredRules and alertsConditions, actionsUpdate orchestrationStaged OTA, rollbackPer-device authentication · multi-tenancy · audit trailApps and dashboardsOperations, customersEnterprise systemsERP, CMMS, CRMAnalytics and AIModels, historySupport and serviceTickets, diagnosticsSecure device link
Reference architecture: devices are identified, authenticated and observed, and their data reaches applications through stable APIs.
Standards and interfaces

What we design against.

MQTT 5 and Sparkplug B
Publish and subscribe messaging for devices, with Sparkplug defining topic structure and state handling for industrial data.
LwM2M and CoAP
Device management and constrained-device protocols for cellular and low-power devices.
OPC UA
Industrial interoperability standard used where devices are controllers and machines.
W3C Web of Things
Thing descriptions that define a device's properties, actions and events in a machine-readable way.
ETSI EN 303 645 and IEC 62443
Security baselines for consumer and industrial connected devices, used as the checklist for identity, updates and data protection.
X.509 and device identity
Certificate-based device identities, with secure elements or trusted platform modules where the hardware supports them.
Data model

The core entities.

Device
Identity, type, firmware, location, owner, tags, state and the credentials it uses.
Device twin
Reported and desired properties, versions and last-seen time.
Telemetry point
Device, metric, value, unit, time and quality, stored by retention policy.
Event and alert
Rule, device, severity, state, acknowledgement and resolution.
Command
Target, payload, status, timeouts and the user who issued it.
Update campaign
Target group, version, stages, success measures and any rollback.
Tenant and entitlement
Customer or unit, devices, users and the data they may access.
Build, buy or extend

When custom development is the right call.

Buy

Cloud IoT services and device management platforms are strong, quick starts for connectivity, ingestion and twins, and the default for most fleets.

Extend

Extend a cloud platform with your own device model, rules, customer-facing applications and enterprise integrations. This is the most common pattern.

Build

Build when you must run in your own environment, when cost per device at scale demands it, when you operate in regulated or air-gapped settings, or when the platform is itself a product.
Security and compliance

Controls designed in, not added later.

  • Strong device identity

    A unique credential per device, protected in hardware where possible, with revocation and rotation.
  • Least-privilege messaging

    Each device can publish and subscribe only to its own topics, and platform components use scoped service accounts.
  • Secure updates

    Signed images, staged rollout, health checks and automatic rollback so a bad update cannot brick a fleet.
  • Data protection

    Encryption in transit and at rest, tenant isolation and retention rules for personal or commercially sensitive data.
  • Fleet monitoring

    Anomaly detection for devices behaving unusually, such as unexpected connections or message volumes.
How we deliver

From discovery to a system you run.

  1. 01

    Fleet and use case review

    We list the devices, the data they produce, the decisions that data supports and the systems that need it.
  2. 02

    Architecture and security design

    Connectivity, identity, device model, storage and integration are agreed with the device, security and application teams.
  3. 03

    Build the core and connect a pilot fleet

    Registry, provisioning, ingestion and twins go first, connected to a small fleet in real conditions.
  4. 04

    Add rules, updates and applications

    Alerts, commands, update orchestration and applications are added and tested at pilot scale.
  5. 05

    Scale the fleet

    The fleet grows in stages, with load tests, monitoring and support processes proven at each step.
Relevant work

Where to see our work.

Related

Keep reading.

Questions

Common questions.

Do you design the devices?
No. We build the platform software and the integration around devices: connectivity, identity, data, twins, updates and applications. Electronics and firmware design stay with your hardware team or vendors, and we agree the interfaces with them.
Should we use a cloud IoT service or build our own platform?
Start with a cloud service unless a requirement rules it out. Building makes sense for air-gapped or regulated environments, very large fleets where unit cost matters, or when the platform is your product.
How do you handle devices that are offline most of the time?
Twins hold the desired state until the device connects, commands have timeouts and retries, and devices buffer data and send it with its original timestamp.
How are firmware updates made safe?
Updates are signed, rolled out in stages to a small group first, monitored with health checks and paused or rolled back automatically if the group misbehaves.
Who owns the code and the data?
You do. The code, the data and any models stay yours.
Next step

Tell us what you need built.

Describe the system and the systems around it. An engineer replies within 1–2 business days.Start a project