PyData Amsterdam 2025, September 24 - 26

Explore the selected talks, deep-dives and tutorials, the full program schedule will be published soon.

Talks

Actionable Techniques for Finding Performance Regressions

Thijs Nieuwdorp

Optimal Observability: Partitioning Data into Time-Series for Enhanced Anomaly Detection and Improved Monitoring Coverage

Vitalie Spinu

Large-Scale Video Intelligence

Irene Donato & Antonino Ingargiola

Scaling Trust: A practical guide on evaluating LLMs and Agents

George Chouliaras

Counting Groceries with Computer Vision: How Picnic Tracks Inventory Automatically

Sven Arends

Is Prompt Engineering Dead? How Auto-Optimization is Changing the Game

Iryna Kondrashchenko & Oleh Kostromin

Uncertainty Unleashed: Wrapping Your Predictions in Honesty

Konstantinos Tsoumas

Causal Inference Framework for incrementality : A Case Study at Booking to estimate incremental CLV due to App installs

Netesh & Nazlı Alagöz

Searching for My Next Chart

Muhammad

Should Captain America Still Host Your Data? A Call for Open, EU-Based Data Platforms

Manuel Spierenburg

Kickstart Your Probabilistic Forecasting with Level Set and Quantile Regression Forests

Inge van den Ende

Resource Monitoring and Optimization with Metaflow

Gergely Daroczi

Formula 1 goes Bayesian: Time Series Decomposition with PyMC

Wesley Boelrijk

Gotta catch ‘em all - Hunting Fraudsters with Minimal Labels and Maximum ML

Jaap Stefels & Itzel Belderbos

Sieves: Plug-and-Play NLP Pipelines With Zero-Shot Models

Raphael Mitsch

Potato breeding using image analysis in a production setting

Dick Abma

What Works: Practical Lessons in Applying Privacy-Enhancing Technologies (PET) in Data Science

Yuliya Sapega

How to Keep Your LLM Chatbots Real: A Metrics Survival Guide

Maria Bader

🍯 Sweet Language Model Python Applications

Jamie Coombes

Declarative Feature Engineering: Bridging Spark and Flink with a Unified DSL

Miguel Leite

Kafka Internals I Wish I Knew Sooner: The Non-Boring Truths

Dima Baranetskyi

Composable Pipelines for ML: Automating Feature Engineering with Hopsworks’ Brewer

Javier

Continuous monitoring of model drift in the financial sector

Denis Gaitan & Agustin Iniguez

Orchestrating success: How Vinted standardizes large-scale, decentralized data pipelines

Rodrigo Loredo & Oscar Ligthart

Deep-dives

Optimize the Right Thing: Cost-Sensitive Classification in Practice

Shimanto Rahman

Model Context Protocol: Principles and Practice

Fabio Lipreri & Gabriele Orlandi

Context is King: Evaluating Long Context vs. RAG for Data Grounding

Bauke Brenninkmeijer

Designing tests for ML libraries – lessons from the wild

Sayak Paul & Benjamin Bossan

Untitled13.ipynb

Vincent D. Warmerdam

Tutorials

AbstractUnderstand your data with Knowledge Graphs

Martin O’Hanlon

Bridging the Gap: Building Robust,
Tool-Integrated LLM Applications with the Model Context Protocol Abstract

Adam Hill & Shourya Sharma

Event-Driven AI Agent Workflows with Dapr

Dana Arsovska & Marc Duiker

Grounding LLMs on Solid Knowledge: Assessing and Improving Knowledge Graph Quality in GraphRAG Applications

Panos Alexopoulos

Meet Docling: The “Pandas” for document AI

Mingxuan Zhao & Michele Dolfi

Listen: A Practical Introduction to Data Sonification

Tomek Roszczynialski & Samuel Janas

Next-Level Retrieval in RAG: Techniques and Tools for Enhanced Performance

Mahima Arora & Aarti Jha