DevOps Foundation - a secure, production-ready operations stack for industrial and IoT platforms

The problem it solves

An IIoT pilot, an energy management dashboard or a line-side application reaches the point where it has to become a real production service. It now needs deployment pipelines, monitoring, alerting, log aggregation, authentication, backups and a tested way to recover. Building that from scratch means hiring a platform team. Adopting Kubernetes means complexity most industrial applications never need.

What it is

A reusable operations backbone, designed and integrated in-house and deployed across multiple live client environments. Built entirely on proven open-source components, so there is no vendor lock-in:

• Container orchestration on Docker Swarm, managed through Portainer

• Traefik as ingress and reverse proxy, with automatic TLS issuing and renewal

• Harbor as a private container registry with vulnerability scanning and image trust policies

• Prometheus, Grafana and Alertmanager for metrics, dashboards and alerting

• Promtail and Loki for centralised log aggregation

• Keycloak and OAuth2-Proxy for single sign-on, MFA and role-based access control

• Firewalled network segmentation with VPN-only access to internal services

• Automated daily and weekly backups of databases and volumes, with tested restore procedures

• Git-driven CI/CD with webhook-triggered rolling deployments and one-step rollback

Why it matters for industrial and energy systems

It runs on-premise as readily as in the cloud, which matters when data residency, plant network isolation or OT separation are constraints - we have built on-premise MQTT infrastructure for exactly that reason. Access is centralised and auditable rather than spread across per-application logins. Container images come only from a scanned internal registry, closing a supply-chain route that is increasingly targeted.

What comes with it

A documented service lifecycle: how a service is onboarded, deployed, monitored, authenticated, backed up, restarted, troubleshot and decommissioned. Our operations team is ITIL-trained, so change management, incident response and disaster recovery follow formal process. Recovery objectives are defined and drilled quarterly: under two hours to restore critical services, under one hour of data loss.

Track record

More than ten production applications currently run on it. Measured availability is 99.95%, with mean time to recovery under fifteen minutes.

Delivery models

We deploy and operate it for you; we deploy it and train your team to run it, with escalation support; or we combine the two.

What are we looking for?

Technology cooperation and commercial agreement with technical assistance. We are looking for industrial, energy and technology companies with a production or near-production system that has no proper operational backbone, and for consortia needing a partner to operate a platform after the pilot phase ends.