DataTalks.Club Meetups Global

How structured retrieval helps high-stakes decision making - Daan Di Scala

​Moving beyond basic RAG isn't just a technical upgrade, it's the key to building truly trustworthy AI. In this session, discover how GraphRAG transforms raw data into reliable insights across three applied real-world use cases.

​In this talk, he will:

• ​Introduce his work at TNO as an AI scientist
• ​Cover the topic of Trustworthy AI decision support, and in particular:
• ​Discuss: how can knowledge-driven RAG and GraphRAG approaches make retrieval more trustworthy?
• ​Take you through 3 applied use cases and showcase the effect of different approaches

​
​About the Speaker:
Daan Di Scala studied Artificial Intelligence at Utrecht University, The Netherlands. After his Master thesis focused on privacy preserving AI agents, he started working at TNO Data Science in 2023 as AI researcher. TNO is the Dutch research institute for applied science, where Daan's research revolves around the trustworthiness of AI in high-stakes domains (agrifood, healthcare, legal). In 2025 he started his PhD at Utrecht University on the topic of Knowledge-Grounded AI for Trustworthy Advice. In his work, he investigates how structured knowledge can enhance state-of-the-art LLM systems to improve the trustworthiness of AI advice, enabling users to better understand, evaluate, and appropriately rely on AI-generated recommendations for high-stakes decisions.

Join our Slack: https://datatalks.club/slack.html

​How to Turn Fragile AI Prototypes Into Customer Ready Systems - Louis-Francois Bouchard

​The transition from a fragile AI prototype to a robust production system requires a completely new set of engineering skills. Modern developers must master prompt engineering retrieval, augmented generation and agentic workflows to build reliable applications. This upcoming episode focuses entirely on the practical realities of shipping generative AI products that actually work for paying customers.

​We will explore the rapidly evolving landscape of AI education and the fascinating rise of vibe coding in everyday software development. The conversation covers how engineering teams can build effective evaluation benchmarks and reduce risks when deploying complex models. We will discover actionable strategies for upskilling workforce and transforming traditional developers into specialized AI professionals ready to tackle real world challenges.

​He'll cover:

• ​Transitioning from academia to building a massive AI education community
• ​Defining the exact skills needed to become a modern AI engineer
• ​Shipping reliable large language models in real world production
• ​Exploring the rise of agentic workflows and vibe coding
• ​Building benchmarks to evaluate the creative writing capabilities of AI

​
​About the Speaker:
​Louis-François Bouchard is the co-founder of Towards AI, where he builds and teaches a practical toolkit for shipping reliable LLM products. He co-authored Building LLMs for Production, a hands-on guide to prompting, fine-tuning, retrieval augmented generation, and evaluation. Through Towards AI Academy, he has launched multiple in-depth courses for AI engineers, designed to turn developers into AI professionals who can transform prototypes into scalable, customer-ready systems. He also runs the What’s AI YouTube channel and newsletter, translating new research and best practices into clear engineering playbooks for 70K+ subscribers and tens of thousands of readers.
​Today, he partners with founders and organizations on AI strategy, training design, and production workflows that raise accuracy, reduce risk, and make generative AI useful for paying customers, and he speaks at events such as AIE, Uphill Conf.

Join our Slack: https://datatalks.club/slack.html