services / legacy modernisation
AI-assisted legacy modernisation – moving legacy systems safely to current technology
Your business-critical software runs on outdated frameworks and no longer receives security updates? I move legacy systems to current technology with AI support – backed by tests, step by step and without putting ongoing operations at risk.
starting point
Typical starting points
- The system runs on outdated frameworks or runtimes that no longer receive security updates.
- Audits or customers demand current, secure technology – but the system is too big for a classic rewrite.
- Knowledge about the software sits in a few heads, and documentation and tests are missing.
- New requirements take forever because nobody dares to touch the old code anymore.
- Developers for the old technology are hard to find.
deliverables
What I actually do
- Assessment – code, dependencies and security risks, considerably faster with AI than by hand
- Documentation of the existing business logic
- Characterisation tests that pin down today's behaviour
- Target architecture and migration plan in realistic steps
- AI-assisted migration to current frameworks and runtimes
- Parallel operation and result comparison before the switch-over
- Handover to your team including CI/CD and enablement for AI-assisted development
Why the calculation has changed
For legacy systems the rule used to be: a rewrite is too risky, too expensive and takes too long. So systems were patched, frozen or pushed to the next budget cycle – while frameworks ran out of support, security gaps stayed open and knowledge about the system shrank with every departure.
AI has changed that calculation. The most laborious steps of a modernisation – understanding code, documenting business logic, writing tests and translating code into a new technology – now take a fraction of the former effort. That makes a rewrite realistic again for many systems.
Experience from mission-critical projects
I have already moved several mission-critical applications to current technology with AI support – systems that urgently needed newer technology and security updates and could not afford to fail. Afterwards they were maintainable, secure and open to further development again.
AI does the heavy lifting. I own the architecture, the review and the quality.
How I work
- Pin down behaviour: before anything changes, automated tests capture what the system does today.
- Switch step by step: module by module following the strangler pattern instead of a risky big bang.
- Compare results: old and new run in parallel where needed until their results match.
- Human review: every AI-generated change is reviewed – nothing reaches production unchecked.
The result: current frameworks and runtimes, closed security gaps and a codebase your team enjoys working on again.
Who I work with
Typical clients are CTOs, heads of IT and engineering managers who are responsible for a business-critical legacy system and need to make it future-proof. In the Hamburg and Rotenburg (Wümme) area I like to start on-site with your team; elsewhere we work mostly remote after a joint onboarding.
stack
Technologies
- Core
- Strong
- Familiar
Architecture & backend
8
- Software architecture – Core
- Domain-Driven Design – Core
- Microservices – Core
- AI-assisted legacy modernisation – Core
- .NET / C# – Strong
- Go – Strong
- Java / Spring Boot – Strong
- Event streaming (Kafka) – Familiar
AI integration
10
- AI agents & tool use – Core
- AI training & enablement – Core
- LLM APIs (Claude, OpenAI) – Core
- Model Context Protocol (MCP) – Core
- Retrieval-Augmented Generation (RAG) – Core
- Evals & quality measurement – Strong
- Cost & latency optimisation – Strong
- Vector search (pgvector) – Strong
- Fine-tuning – Familiar
- Self-hosted / on-prem models – Familiar
process
How we work together
Arrive on-site first, then deliver efficiently remote – anywhere in Germany and the DACH region.
- 01
Intro call
30 minutes, free – goals, scope and whether we're a good fit.
- 02
On-site onboarding
A few days with your team to learn people, systems and domain.
- 03
Remote delivery
Mostly remote, with fixed rituals and visits as needed.
- 04
Handover & enablement
Documentation and knowledge transfer – your team takes over.
faq
Frequently asked questions
Is it worth rewriting our legacy system?
Far more often than it used to be. With AI support, even large codebases can be moved to current technology in a manageable time. What matters is a controlled approach: tests pin down the existing behaviour, the switch happens step by step, and the system stays in operation throughout.
Isn't AI-generated code risky?
Only without control. I review every change, automated tests pin down the behaviour, and before the switch-over we compare old and new. AI speeds up the work – the responsibility stays with a human.
Which technologies do you modernise?
My focus is .NET, Java with Spring Boot, TypeScript/Node.js and Go – as source and as target technology. What matters more than the language is a clean, testable approach.
Can our source code be shared with an AI at all?
We clarify that together up front. Depending on your requirements I work with providers that do not train on your data and offer a data processing agreement, or with self-hosted models.
How long does a modernisation take?
That depends on the size and state of the system. After the assessment you get a realistic plan with milestones. Because the switch happens step by step, the first modernised parts go live early – not only at the end.
contact
Let's talk about your project.
Tell me briefly what it's about – I'll get back to you within 1–2 business days.