Workshops: Agentic Coding
Modernising Legacy Code with AI Agents: Workshop
Old codebases without tests or documentation are an ideal yet risky place for AI agents. In this two-day workshop your team learns to understand legacy systems with agents, lock in behaviour with tests and modernise them step by step without putting the running business at risk.
Who is this workshop for?
- Teams maintaining long-lived legacy systems
- Software architects tasked with modernisation
- Developers taking over unfamiliar code
- IT leaders with technical debt in their portfolio
- Teams facing framework or language migrations
What you will take away
- Have agents analyse and document unfamiliar code and surface dependencies.
- Lock in existing behaviour with characterization tests before changing anything.
- Refactor in small, verifiable steps with the agent.
- Plan and start framework upgrades or language migrations with agent support.
- Apply patterns such as the strangler fig to replace legacy systems gradually.
Agenda
Module 1
Day 1: Understand
- Exploring the codebase with Claude Code, OpenAI Codex or Cursor
- Generating architecture and dependency overviews
- Creating missing documentation: modules, data flows, business rules
Module 2
Day 1: Secure
- Generating characterization and golden master tests
- Making code testable: seams and breaking dependencies
- Exercise on your legacy module
Module 3
Day 2: Refactor
- Small steps, frequent commits, tests as safety net
- Reducing dead code, duplication and complexity
- Review strategy for large diffs
Module 4
Day 2: Migrate
- Framework upgrades, e.g. older Java, .NET, PHP or AngularJS versions
- Language migration and translating business logic
- Strangler fig and parallel validation
Module 5
Day 2: Roadmap
- A modernisation plan for your system
- Risk assessment and rollback strategy
- How agents support ongoing operations
Why agents are so effective on legacy code
An agent reads thousands of lines of unfamiliar code without fatigue, traces call chains and explains business logic in plain language. For teams that inherited a system, this shortens onboarding considerably.
The danger is overconfidence: an agent rewriting unknown code without tests can introduce subtle behaviour changes. So the workshop follows a strict order: understand, secure, change.
Typical modernisation scenarios
We align the exercises with your real system. Common scenarios include:
- Upgrading outdated framework versions
- Splitting a monolith into modules or services
- Replacing unsupported libraries
- Migrating between languages or platforms
- Adding test coverage to critical modules
Format
Two days, hands-on with your code, delivered by neooutput AI Solutions Architects and AI Code Engineers. Remote or on-site, in English or German. jocapps is ISO 27001 certified; handling of your source code is agreed contractually in advance.
Frequently asked questions
Does this work with very old languages?
Modern models often understand older languages and frameworks well. How far a migration can be automated depends heavily on the code; we assess this together on your system.
Can our source code go to AI providers?
We clarify that upfront. With enterprise plans that exclude training use and suitable contracts, many organisations can. Alternatively we work with a representative excerpt.
Is the goal a fully automated migration?
No. The goal is a stepwise, safeguarded modernisation where agents do the heavy lifting and your team makes the decisions.
Workshop
Get in touch
Which legacy system should be modernised first?
Tell us a little about your team – we will get back to you within 24 hours with a proposal for content, format and dates.
Prefer to talk?
More workshops
- Workshops: Agentic CodingHow to Review AI-Generated Code Securely: Workshop
- Workshops: Agentic CodingClaude Code for Non-Developers: Workshop for Product, Design & Ops
- Workshops: Agentic CodingTrain the Trainer: Scaling Agentic Coding Across Your Company
- Deploy, Convert & ScaleScaling a Windsurf-Built App: Performance, Security and Teams
