Munich Datageeks November Edition

We are thrilled to announce our next Meetup on November 25th at E.ON.
Format:

• 2 talks (each ca. 40 min incl. discussion)
• Time for networking + food + drinks before, in between, and after the presentations
• Talks are held in English
• We will be taking photos and/or film footage at the event. These will be used to share news about our meetups and to publicize upcoming events.

The lineup:
First talk:
Alisa Bogatinovski - Beyond Accuracy: Rethinking Evaluation for LLM Classifiers

Abstract:
LLM classifiers often appear straightforward, yet ambiguity, inconsistent outputs, and shifting definitions make their evaluation surprisingly difficult. Standard metrics rarely capture the full picture, especially when real-world cost and operational impact matter. This talk outlines practical methods for evaluating LLM classifiers beyond accuracy, combining curated datasets, semantic assessment techniques, and feedback from human interactions to build systems that remain reliable over time.

Bio:
Alisa Bogatinovski is a Senior Data & AI Scientist at E.ON Digital Technology, specializing in GenAI applications for the energy retail domain. With a background in computer science and experience across multiple data science roles, she focuses on applied AI systems, including NLP, agentic workflows, and multimodal processing.

Second talk:
Emanuel Sommer - Sampling: The future of Bayesian Deep Learning?

Abstract:
Sampling in Bayesian neural networks has long carried a reputation for being elegant yet impractical for large-scale or complex models. But times have changed. Recent progress in hybrid and scalable samplers, as well as software, is reshaping what’s possible for Bayesian neural networks. This talk highlights how these methods achieve both reliable uncertainty estimates and competitive predictive performance, challenging long-held beliefs about the limits of sampling.

Bio:
Emanuel Sommer is a Doctoral researcher in Statistics and Machine Learning at the [Munich Uncertainty Quantification AI Lab MUNIQ.AI (https://www.muniq.ai/) at LMU Munich and a member of the Munich Center for Machine Learning (MCML) (https://mcml.ai/). His research focuses on sampling-based inference for Bayesian Neural Networks, aiming to bridge the gap between theoretical rigor and practical applicability in probabilistic deep learning. Before his PhD, Emanuel earned his B.Sc. and M.Sc. in Mathematics from TUM and gained industry experience as a Data Scientist.