AI & Agentic AI consulting

Before a company invests in agentic AI, it needs clarity on where that investment actually pays off and where a simpler system or a rule-based process would do the job just as well. That's where this consulting starts: I assess use cases for technical feasibility, data availability, and business value before a single line of code gets written.

A typical engagement begins with conversations with subject-matter experts and decision-makers to understand existing processes and pain points. The result is a prioritized roadmap: which use cases are a good fit for RAG pipelines, where agents can meaningfully automate work, and which privacy, security, and operational requirements need to be considered from day one.

This assessment is grounded in direct project experience. For Silver Investment Partners, I evaluated how retrieval-augmented generation could make financial and operational metrics extracted from information memoranda comparable, and built the investment analysis platform around that assessment. For COMAN Software GmbH, I assessed the data-protection requirements and translated them into a fully on-premises architecture running local LLMs, where no sensitive data goes to external cloud services. And for leezy.ai, I made the early architectural calls on system design, RAG pipelines, and prompt engineering strategy that now carry a production, scalable customer service agent.

The outcome is never a generic technology recommendation. It's a roadmap that fits the data, the team, and the budget at hand.