— AI & Consulting

AI and consulting

Alongside teaching, I have spent more than twenty years as a network architect, with a focus since 2024 on generative-AI systems. If you have a concrete problem you want to solve with modern AI tooling, get in touch.

— What I offer

MCP servers and agents

Production-grade Model Context Protocol servers that expose external systems — networks, databases, APIs — to LLM agents. I have built a 90-tool MCP server for Cisco NSO and a 45-tool server for Cisco WAE.

Agentic workflows with Cursor

Structured agent pipelines using the Plan → Execute → Verify → Iterate pattern. Domain-specific Cursor rules, skills and anti-hallucination frameworks for real engineering tasks.

Transformer fine-tuning

QLoRA fine-tuning of 7B-parameter models (StarCoder2, Qwen2.5) on bespoke datasets, using PyTorch, TRL and PEFT on GPU hardware.

RAG and graph-based systems

Retrieval-augmented-generation pipelines built on LlamaIndex and vector databases; graph-based modules for structured analysis.

— Tools & stack

Tools & stack

Python · FastMCP · LlamaIndex · PyTorch · TRL/PEFT/QLoRA · Hugging Face · Cursor (Agent Mode) · MCP Protocol · Git/GitHub · SSH · Conda. On the network side: Cisco NSO 6.4, RESTCONF, IOS-XR, YANG, WAE, SR-TE.

— How I work

How I work

Start small, verify early, roll back often. No hype demos — I build systems that run in production and leave behind code, tests and documentation.

— FAQ

Frequently asked questions

What is an MCP server and when do I need one?
MCP (Model Context Protocol) is an open standard that lets LLM agents — Claude, Cursor, custom assistants — talk safely to external systems: databases, APIs, network tooling, your own software. If you have a workflow where a colleague runs Claude Code or Cursor and needs to reach internal infrastructure, a dedicated MCP server is usually the right architecture.
Do you work remote, or only in Bavaria?
I work AI engagements largely remote — that is the industry norm. Workshops and longer on-site engagements I prefer to do around Ingolstadt, Munich, Augsburg and Bavaria. Germany-wide by agreement.
What specific generative-AI experience do you have?
Full-time since 2024. Production MCP servers for Cisco NSO (90 tools) and Cisco WAE (45 tools). QLoRA fine-tuning of 7-billion-parameter models (StarCoder2, Qwen2.5) on bespoke data. RAG pipelines with LlamaIndex. Agentic workflows with Cursor. Before that, 20+ years as a CCIE/CCDE network architect at Cisco.
What does AI consulting cost?
Hourly and day rates by arrangement — depending on scope, complexity, and travel. An initial scoping engagement (typically half a day) can often be flat-rated. Send a short description of what you have in mind and I will reply with an estimate.

Send an inquiry

Briefly describe your problem or project in the contact form. I will reply with an assessment.

Go to the contact form →