Companies are producing and have to manage more and more information. Employees waste valuable time searching for documents instead of solving problems.

As the founder of Heureka AI I am working on the AI powered Knowledge Management System named Pallas. Pallas is build for SME’s and enables employees to access and query all company internal documentation and knowledge at a single point.

Combining LLM’s with a Retriever Augmented Generation (RAG), the system is able to analyzes multiple different Knowledge Bases like Confluence, Web, Sharepoint, Dropbox. This knowledge is then available to be found through semantic search and a Question/Answer chat system.

With Pallas companies can manage the information overload and fragmentation, making sure their employees waste no time searching for information but use it to solve problems.

Customers are talking to service agents every day, but its difficult to capture what customers care about and what the struggle with.

As Lead Engineer I was building a Customer Analytics Platform for Cariad that was analyzing customer service calls to extract insights about customer complains and opinions about VM vehicles and services.

The system analyzes all German Service Center conversations, extracting the conversation topic, mentioned products and services, as well as sentiment progression in order to identify issues and provide implicit customer product satisfaction and feedback.

This makes it possible for branch managers to decide how to develop new services and products based on customer insights.

Everyone is talking about AI, but very few know how it works. Employees are interested and worried alike, fearing about their jobs and on missing out.

I provide extensive in-house Lecture and Training on the history and state of the art of Large Language Models like ChatGPT. Helping employees to understand how AI can benefit them, and helping the company to identify and select first AI projects as well to identify challenges and opportunities in their implementation.

This reduces anxiety and enables the company to make use of the creativity and ingenuity of their people.

Implementing continuous training of Machine Learning Models is hard. Keeping track how and on what data the latest model was trained is even harder.

As AI Consultant I designed a MLOps Architecture to track the training data sets used for image analysis pipelines. Enables training data lineage, which describes what training data was used to train which model and is an integral part in automatic re-training Machine Learning Pipelines to ensure quality and stability of the system.

This provides control and ensures robustness. Making it possible to understand where changes in the behavior or performance of models come from and where to start investigating if problems emerge.

Modern factories are requiring thousands of spare parts to ensure optimal production. It is essential to have an overview of what parts are in use, if they can be replaced and how many one should have in stock.

As Data Engineer and Lead Developer I am building a Spare part Management System for Factories, that keeps track of current and historic spare part usage.

Up to date supply chain information, combined with a history based demand forecast, enables Factory owners to optimize stock levels in order to reduce costs, while preventing the interruption of operations because of supply chain interruptions.

This enables factory floor managers to optimize costs and ensure the uninterrupted production.