Notes from projects, teaching and research, no marketing buzz.
The uncomfortable truth from almost every project: the model is rarely the problem. How bad data ruins good AI, and what helps.
What retrieval-augmented generation does, why an agent goes further, and where the limits remain.
On 18 June 2026, the B2B innovation network meets in our home town, Bielefeld. We will be there as exhibitors.
Fully automatic or with a sign-off? When writing with generative AI, the right division of labour decides quality and trust.
We are very happy to be an exhibitor at OMR in Hamburg. You will find us in the Premium Hall at B6 I07.
What robots.txt really is, what its RFC says, and why assuming everyone obeys it stays wishful thinking.
Our lead engineer Sergej Schultenkämper has completed his PhD with the distinction summa cum laude. We congratulate him wholeheartedly.
We are delighted that Dr Frederik Simon Bäumer is now professor of information systems with a focus on data science at HSBI.
A small blind test with five English translations of one German consulting sentence, and why corporate language wins in the end.
Why ChatGPT & co. are genuinely useful, but don't replace a search engine, and what we do at wonk.ai instead.
