ALL SPEAKERS
SESSIONS
SCHEDULE TBD
KEYNOTE
Perfect queries, overconfident conclusions
AT A GLANCE
Role
Head of Data Science & ML
Organisation
Deichmann
Category
Partner
ABOUT THE SPEAKER
Julian's team of data scientists and ML engineers covers everything from reporting to analytics products for Europe's largest footwear retailer, including demand forecasting, pricing, customer analytics and digital analytics across store and e-commerce. That spans brands such as Deichmann, Snipes and Ochsner Sport in more than 30 countries. He came to the field through experimental particle physics and worked as a data scientist in manufacturing at thyssenkrupp before joining Deichmann.
TALK
Perfect queries, overconfident conclusions
Generative AI made it dramatically cheaper to produce a pipeline, a query, a dashboard or an analysis. It did not make it cheaper to trust the result. In a company that has been selling shoes for more than a century and, more recently, bikes and skis, source systems change constantly, a size does not mean the same thing across brands, and half the meaning of a column lives nowhere in the schema. A model writes solid SQL and Python. Syntactically the query is perfect. It cannot know what those columns means. That is the kind of thing everybody knows and nobody writes down. Notes and examples from a data organisation working through this, across pipelines, analytics, forecasting and the business acting on the numbers: what genuinely got faster, what breaks quietly, and why supplying context turns out to be the hard part, for models and for people alike.
MORE VOICES
Other speakers at code.talks
See them live — and 100+ more.
One ticket, every track. Nov 4–5, 2026 at Kinopolis Hamburg.
Get your ticket





