AI only creates dependable value once it is embedded in a working process.
I develop AI-powered processes and software products together with companies. In doing so, I connect AI models, classic automation, existing systems, and human decisions into a single end-to-end workflow.
The engagement can start with a focused prototype, a production-ready first release, or the development of a custom AI system.
Test a concrete idea in practice
An AI prototype tests one central assumption with an executable workflow and realistic data.
In a joint workshop we narrow down the use case, define the intended outcome, and identify the decisive technical and domain uncertainties. Immediately afterwards, I build a first working slice of the process.
From workflow to a production-ready first release
I take a clearly bounded use case to a production-ready first release within four weeks.
The workshop is not an upfront strategy phase. It is the start of the joint development. Once process, scope, and success criteria are defined, technical implementation begins immediately.
Developing more complex processes, products, and platforms
Some initiatives cannot reasonably be reduced to a single process or a compact first release.
I develop custom AI systems for initiatives with multiple processes, integrations, user groups, or elevated technical and organizational requirements.
You receive a working prototype that lets you judge in practice:
The exact price depends on the data situation, the integrations involved, and the size of the process segment under test.
You receive a production-ready first release for a clearly defined use case. Production-ready means the agreed core process:
Delivery within four weeks requires that the use case is sufficiently bounded, the necessary data and access are available, and decisions can be made quickly during development.
The result is a software system tailored to its specific purpose, bringing together processes, data, AI, automation, and human work.
Job Directory is an AI-powered process product I built for structured job searching and application preparation. The system connects several components into one continuous human-AI process:
Humans stay responsible for the decisions that matter. The AI takes over research, preparation, scoring, and groundwork.
The case shows how a recurring, information-heavy workflow can become a standalone product with automation, data storage, a user interface, and human control points.
The offerings are a good fit for work such as:
Not every process needs AI. Where rules, classic software, or existing automation tools are the better solution, that is what gets used.
In a first conversation we clarify the current process, the intended outcome, and the key constraints.
Based on size and uncertainty, we classify the initiative as an AI prototype, an AI product, or a custom AI system.
Before the engagement starts, we define the concrete deliverables, necessary integrations, success criteria, and your part in the work.
Workshop and implementation are tightly linked. Insights from development feed directly into the next step.
The system is tested with realistic cases, documented, and made available for the agreed use.
After the first release, the product can be extended and improved in further development iterations. Possible next steps include:
Every further iteration gets its own scope and a defined outcome.
You have a manual, information-heavy, or recurring process where AI could make a practical difference? Briefly describe:
We can then determine whether a prototype, a production-ready first release, or a custom system development is the most sensible way to start.
Describe it in a few sentences. I will get back to you within one business day with a candid view on the most sensible way to start.