Steven Reinisch

Do you want to do more than just try out AI agents, but integrate them meaningfully into your business processes?

I support start-ups, SMEs, and corporations in the technical management of AI agent projects: from feasibility studies, architecture and provider decisions, to implementation and rollout. My focus is on pragmatic solutions with measurable benefits, rather than AI demos without connection to everyday work.

For over 2 years, I have been continuously working with teams at Amazon Robotics to automate digital work with AI agents. This includes error analysis and resolution in the installation and operation of industrial robots, automated follow-up communication with project participants, as well as reporting and operational project support.

The results show what is possible: In an agent project for robot error analysis, the MTTR – Mean Time to Recover – was reduced by 36%. A follow-up agent saves project managers approx. 4 hours per month, and automated reports save an additional approx. 3 hours. In larger teams, this effect scales accordingly.

I combine technical depth with project management experience: I can assess what is realistic with Claude Code, OpenAI Codex/ChatGPT, or individual agent implementations, which architecture makes sense, and which risks need to be considered regarding data, compliance, operations, and costs.

A typical starting point is a fixed-price feasibility sprint. In this sprint, I analyze your processes, check data and compliance requirements, compare provider and architecture options, identify risks, estimate effort and costs, and provide a clear recommendation: Build, Buy, Prototype, or do not proceed for now.

If the results are promising, I can then coordinate the technical implementation and guide your AI agent project from idea to productive use.