I turn ambiguous domain and operational requirements into reliable software systems – from discovery and domain modelling through implementation, rollout, verification and operational follow-up.
I have worked as a software engineer for more than 30 years, building complex, long-lived systems. My foundation is backend and API architecture, data modelling, integration and platform operations. My current focus is AI and platform systems in real-world environments: understanding domain and operational constraints, making risks visible and delivering solutions that remain accountable in production.
I do not treat AI primarily as a code generator. I am interested in embedding AI agents into controlled engineering workflows with explicit target states, review, validation, audit trails, rollback boundaries and human ownership.
Selected Evidence
- Rebuilt and integrated a complex intellectual-property import module and maintained APIs connecting legacy systems, customer-facing integrations and newer internal systems.
- Delivered custom ASP.NET/SQL Server applications in chemical and pharmaceutical environments covering production support, laboratory workflows, IT asset management and regulatory reporting.
- Built an ISO 9001 / GAMP 5 software-development QMS with risk analysis, change control, traceability, validation artifacts and developer-facing SOPs as a foundation for accountable AI-assisted SDLC workflows.
- Designed optimisation and data-processing software for logistics networks, including NP-hard line-haul optimisation, OSM/OSRM tooling and geocoding workflows for multi-million-row datasets.
- Operate a production-style RKE2/GitLab platform as a real-world lab for Git-based delivery, infrastructure automation, recovery and controlled agentic workflows.
For technical detail, see my blog, especially GitLab as an Operating System for AI-Assisted Infrastructure. Code and additional projects: GitHub.