AI Engineering &
LLM Integration.
Most recently I led frontend architecture and CMS modernization for lastminute.com, driving AI-augmented development practices across large-scale refactors.
I am now focused on evaluating models via vLLM, Ollama, and LM Studio. I am architecting local AI infrastructure, RAG pipelines combining vector search with local/cloud LLMs, and workflow automation tooling using n8n and Model Context Protocol (MCP) servers.
Local AI Infrastructure Platform
Architecting multi-GPU local AI development environments for LLM experimentation. Focusing on inference performance and cost efficiency.
RAG Pipeline & Knowledge Systems
Building retrieval-augmented generation pipelines. Experimenting with chunking strategies and hybrid search architectures.
AI Workflow Automation
Designing automated workflows using n8n integrated with MCP servers for task orchestration.