Holistic protection and seamless traceability for cybersecurity, functional safety, and the Cyber Resilience Act
Overview
Methodological excellence and tailored tools for model-based system and software engineering.
Overview
Enterprise software from a single source: AI integration, legacy migration and full-stack development — cost-efficiently and sovereignly hosted.
Tailored software development for enterprise customers — full-stack, AI-assisted and agile. For CTOs, IT directors and enterprise architects in banking, insurance, logistics, energy, public sector and industry.
Custom software development is more than delivering features. It requires the interplay of a well-considered engineering strategy, the right integration of AI and the ability to modernise legacy systems step by step — without putting ongoing operations at risk.
itemis combines three disciplines for exactly this: AI Enablement for the sustainable use of AI in development and organisation, Legacy Modernization for the methodical, incremental migration of legacy systems with provable equivalence, Full-Stack & Cloud for cloud-native platforms in regulated industries. Senior teams with a strong open-source tradition — and the conviction that quality is non-negotiable.
AI in the development process is more than a Copilot subscription. Sustainable adoption means: analysing processes, selecting the right tools, ensuring data sovereignty, and involving the people in the organisation from day one.
Here you will learn how AI Enablement differs from coaching and tool sales, why data sovereignty does not have to be a blocking issue, what the path from a first pilot to independent operation looks like, and when we advise against AI.
Legacy code gets more expensive every year, and simultaneously harder to modernise, because the know-how leaves the company along with the developers. The way forward is a methodical, incremental approach.
Here you will learn why the strangler-fig pattern is structurally superior to big-bang rewrites, how the five-stage model guarantees safety at every level, what role AI-assisted analysis toolchains play, and what a concrete project blueprint looks like.
Full-stack means: frontend, backend, API design, data modelling, deployment, observability. A coherent engineering discipline, not a collection of sub-contractors. For regulated industries, sovereign cloud and compliance requirements are part of the equation.
Here you will learn how senior teams deploy AI toolchains productively through advanced context engineering, which enterprise stack we use, what sovereign cloud and Gaia-X-compliant architectures mean for regulated industries, and what a structured project onboarding looks like.
Topics at a glance
Schedule a call with Holger Schill and Christoph Hess.
We staff projects with cohesive teams and engineers who are motivated because we take them seriously. Clients notice the difference from the very first sprint.
Audits, release weeks, crisis sprints — we stay calm, maintain focus and deliver under pressure every time. Reliability that goes beyond the contract.
No diplomatic detours, no politics. We communicate directly, listen carefully and speak plainly when something isn't right. A sparring partner, not a supplier.
We invest in the partnership, not the sales pitch. Those who know us know: at itemis, reliability is not something we promise — it is something we demonstrate.
Compare systems, not models. Whether a self-hosted open-weight model holds up is decided by the context you assemble, the tools it can call, the checks it has to pass and the training data your workflow produces on its own. Christoph Hess sets out five engineering levers and the conditions under which each one pays off.
Download Whitepaper
Evaluating Your Migration Options — a joint session with itemis and Haulmont on life after Camunda 7's end of life.
Watch Webinar
How to hand the bulk of software development to AI agents – and still get a result you can trust. A practitioner's look at vertical agents and horizontal skills for enterprise software teams.
Read Article
Arne Deutsch
8 min read
AI-driven development rarely fails because of the code – it fails because of aimlessness. Why technology is only half the battle and what really determines project success.
Read Article
Holger Schill
4 min read
Camunda 7 CE is reaching end-of-life. Operaton and the OpenBPM Platform offer a powerful open alternative – with minimal migration effort and maximum future-proofing.
Read Article
Karsten Thoms
2 min read
Lessons learned from migrating itemis CREATE, a 280K LOC Eclipse-based application, to the web and Visual Studio Code over two years.
Read Article
Andreas Mülder
10 min readWith React Native, itemis delivered one codebase for iOS and Android to connect TECE's hygiene flushes via Bluetooth.
View ProjectREMONDIS streamlined sales by replacing Word macros with a mobile app for offer and contract creation.
View ProjectThalia modernized its eCommerce platform from a legacy monolith to self-contained systems with itemis.
View ProjectBLG LOGISTICS cut heating costs and reduced staff strain with automatic gate control via a smartphone app.
View ProjectCustom software pays off when core processes are too specific for a standard product, when competitive advantages lie in proprietary logic, or when integration requirements exceed what a standard product can handle. Standard products are the better choice for peripheral processes with no differentiation requirements.
The deciding factor is not cost but the strategic significance of the software to the business model. Custom development is an investment in core competency — not a substitute for missing process discipline.
AI accelerates standardisable tasks: code reviews, test generation, documentation, routine refactorings. What AI structurally cannot do: make architecture decisions with long-term consequences, translate domain knowledge into data models, or meaningfully interpret requirements in regulated environments.
The most valuable use of AI in software development is not replacing engineers but taking over routine work — so more capacity remains for decisions where context and judgement matter.
Big-bang migrations frequently fail because complexity is underestimated and operations cannot be interrupted. The more reliable approach is incremental migration following the Strangler Fig pattern: new functionality is built alongside the legacy system, existing modules are replaced step by step until the legacy system is fully retired.
The prerequisite is a stable interface design between old and new, with clear migration paths for business processes and data — and defined intermediate states that remain usable in production.
Fast feature delivery maximises short-term deliverable functionality — often at the expense of architecture quality, test coverage and maintainability. That creates technical debt which shows up as rising change costs, fragile releases and hard-to-predict future development.
Sustainable engineering makes deliberate decisions that are difficult to undo later: database design, API interface stability, deployment architecture, test strategy. This does not mean slower delivery — it means delivery speed that does not come at the cost of the next three years.
