We are thrilled to announce Pydata’s next meetup on Thursday, April 30th, 2026, hosted by us, Artefact!
This edition dives deep into the world of conversational AI and agentic workflows. We will explore how organizations are moving past the prototype stage to deploy LLM-powered systems that actually hold up in production across analytics, logistics, and retail.
Whether you are an AI engineer, a Data Scientist, or a builder curious about what it takes to bring these systems to life, you’ll walk away with a clearer picture of the real challenges behind scaling conversational AI.
Getting an agent to work in a demo is one thing. Getting it to work reliably with real users, real data, and real constraints is a different problem entirely. In this edition, we are peeling back the layers of production-grade AI, looking at the architectural decisions, the retrieval strategies, the guardrails, and the trade-offs that separate a promising proof-of-concept from a system that actually delivers.
This evening goes beyond the basics of prompt engineering and API calls. We are diving into how teams are building context-aware agents that help logistics employees onboard faster, guide retail customers through complex decisions, and unlock self-serve analytics at scale. You will discover what these systems look like under the hood, and what it really takes to ship them.
Excited as well?! We would love to welcome you for an evening full of knowledge sharing, deep technical dives, and of course great conversations, networking, and a fun evening with the community!
Agenda
-
17:30 – 18:25: Welcome with food and drinks!
-
18:25 – 18:30: Artefact Intro
-
18:30 – 19:10: [Talk 1] Agents… in the data-verse! by Adithya Krishnan
-
19:15 – 19:30: Break
-
19:30 – 19:50: [Talk 2] : Building Active Assistance and Agentic Workflows for Logistics by Diederik Heijbroek
-
19:50 – 20:10: [Talk 3] : Conversation agents for Retail by Lorenzo Casimo
-
20:10 – 21:00: Networking / drinks
Talk 1: Agents… in the data-verse! Building the systems that will enable self-serve analytics! by Adithya Krishnan
Self-serve analytics sounds simple—until you look under the hood. Adithya will share what it really takes to build the infrastructure that lets agents query, reason, and respond at scale.
Talk 2: Building Active Assistance and Agentic Workflows for Logistics by Diederik Heijbroek
In the fast-paced world of logistics, employee efficiency and onboarding are critical operational bottlenecks. This talk explores a real-world Agentic AI use case, detailing how we built and deployed an agentic workflow using LangChain to tackle these challenges.
Talk 3: Conversation agents for Retail by Lorenzo Casimo
A leading Danish beauty and wellness retailer partnered with our team to design and deploy a conversational AI agent transforming how customers discover products and resolve beauty concerns online. The solution combines semantic product search, knowledge retrieval, and a guided problem-solving flow to replicate the in-store advisor experience at scale. Built with compliance and brand safety at its core, the architecture balanced multi-turn context, real-time guardrails, and structured user interaction across web and mobile. This case study explores the agentic design decisions, feature trade-offs, and success metrics behind bringing conversational retail AI from concept to production.
About Pydata
PyData Amsterdam is a vibrant community of Python and data enthusiasts that brings together this community and provides a forum for users and developers of open-source data tools.
Speaker(s)
Diederik Heijbroek, Senior Machine Learning Engineer
Artefact
Diederik specializes in designing and building production-grade AI systems around large language models, from Retrieval-Augmented Generation pipelines to Knowledge Graphs. Before joining Artefact in early 2025, he worked as a researcher analyzing keystroke dynamics as early predictors of neurodegenerative diseases. Now at Artefact, he leverages this foundation to transform complex technical architectures into scalable AI integrations for business processes.
Lorenzo Casimo, Senior Data Scientist
Artefact
Lorenzo applies machine learning and large language models to solve real-world client challenges. Before joining Artefact in 2025, he spent over two years at MSCI working on the ESG desk. With an MSc in Environmental Data Science and Machine Learning from Imperial College London, Lorenzo brings a rigorous quantitative foundation to problems that span from predictive modeling to production-grade AI systems.
Adithya Krishnan, Software Engineer
MotherDuck
Adithya Krishnan is a software engineer at MotherDuck, working on AI-related features, from UI/UX to in-database LLM capabilities. His interest in building interactive applications started with his first project, Greppo, which made him realize how much he enjoys creating frameworks and tools, especially those involving AI/ML.
