Thinking of contributing to open source but don’t know where to start? At #PDAmsterdam2024, Magdalena Kowalczuk shared what it’s really like to take that first step. 🎥 youtu.be/IY9Dl2gD_Rw 📅 Join us at #PDAmsterdam2025 (Sept 24–26): pretix.eu/pydata/amsterd…

📣 The selected tutorial sessions are now live! Check out the list here: amsterdam.pydata.org/program 🎟️ Super early bird tickets are still available, don’t miss out: amsterdam.pydata.org/tickets #PDAmsterdam2025

At #PDAmsterdam2024, Ankur Ankan introduced causal inference using DAGs and pgmpy, showing how to move from correlation to causation with real data and accessible tools. 🎥 youtu.be/uSkmnBEAvN8 📅 Join us at #PDAmsterdam2025, Sept 24–26: pretix.eu/pydata/amsterd…

We’re excited to welcome Xomnia as a Platinum X Sponsor of PyData Amsterdam 2025. Their thoughtful and hands-on approach to data and AI continues to make a meaningful difference, and we’re glad to have them as part of the community this year. 🔗 amsterdam.pydata.org/tickets

🎉 We're excited to announce Dr. Maria Börner as a keynote speaker at PyData Amsterdam 2025, celebrating our 10-year anniversary! 🇳🇱✨Don’t miss the chance to hear her speak this September in Amsterdam, and celebrate a decade of PyData Amsterdam with us! amsterdam.pydata.org/conference

We’re excited to welcome Demetrios as one of our keynote speakers for this year’s PyData Amsterdam, happening September 24 - 26. join us this September: amsterdam.pydata.org/conference

📣 Join us on July 15 for our next meetup, co-hosted with PyLadies AMS , at PLAYERUNKNOWN Productions ! We'll explore how AI is transforming game development, from landscape generation and “sentient” NPCs to inclusive design. 🔗lnkd.in/eNeyaRRd

The PyData Amsterdam 2025 Program is LIVE! 🎉 We’re excited to unveil the program schedule for this year’s #PDAmsterdam2025. 🔗 Explore the talks, deep-dives keynotes and tutorials here: amsterdam.pydata.org/program

In the talk 🏎️ "Formula 1 Goes Bayesian: Time Series Decomposition with PyMC," Wesley Boelrijk demonstrates how to build interpretable time series decomposition models using PyMC, a cutting-edge probabilistic programming library. 🔗Full program: amsterdam.pydata.org/program

At #PDAmsterdam2025, Rodrigo Loredo & Oscar Ligthart from Vinted, will present the architectural and practical decisions behind Vinted's approach to managing large-scale, decentralized data pipelines. 🔗 Explore the full program: amsterdam.pydata.org/program

Building language model–powered apps can get messy fast, but it doesn't have to be. In "🍯 Sweet Language Model Python Applications," Jamie Coombes shares how to turn those sticky situations into smooth, scalable solutions. Explore the full program: amsterdam.pydata.org/program

Is prompt engineering dead? In “How Auto-Optimization is Changing the Game,” Iryna Kondrashchenko & Oleh Kostromin explore tools like DSPy and MIPRO that are redefining how we craft prompts. Explore the full program: amsterdam.pydata.org/program

Are cloud costs creeping up? In “Resource Monitoring and Optimization with Metaflow,” Gergely Daroczi shares how to track resource usage and cut waste with a simple, powerful extension. Explore the full program: amsterdam.pydata.org/program

In the talk "Actionable Techniques for Finding Performance Regressions",Thijs Nieuwdorp & Jeroen Janssens share how they uncovered and fixed a subtle performance issue in the Polars DataFrame library. Explore the full program: amsterdam.pydata.org/program

In "Uncertainty Unleashed: Wrapping Your Predictions in Honesty," Konstantinos Tsoumas breaks down how conformal methods can bring calibrated, trustworthy uncertainty estimates to your models. 🔗 Explore the full program: amsterdam.pydata.org/program

In"Gotta Catch ’Em All-Hunting Fraudsters with Minimal Labels and Maximum ML,"Jaap&Itzel shares how their team built a production-ready ML system that processes hundreds of transactions per second, all w/out relying on confirmed fraud labels. 🔗amsterdam.pydata.org/program

Worried your chatbot is hallucinating? In “How to Keep Your LLM Chatbots Real,” Maria Bader shares practical metrics and tools to evaluate and improve GenAI apps. Explore the full program: amsterdam.pydata.org/program

Not all model mistakes are equal. In “Cost-Sensitive Classification in Practice,” Shimanto Rahman shows how to train models that optimize for real-world impact using the Empulse library. Explore the full program: amsterdam.pydata.org/program

LLMs are only as useful as the context they access. In “Model Context Protocol,” Gabriele Orlandi & Fabio Lipreri share how MCP connects tools like GitHub & Slack to your agents. Explore the full program: amsterdam.pydata.org/program