Context for AI | Leiden AI Community Meetup

Context is becoming the missing link that transforms powerful AI models into truly intelligent systems.
In 2026, the next leap in AI comes from understanding context - user intent, environment, history, and goals. This shift separates generic outputs from meaningful, real-world outcomes.

Join this Meetup to explore why context is the foundation of more reliable, human-centered, and transformative AI. We’ll dive into the role of knowledge graphs and how they help AI systems make sense of complex information and connect insights the way humans do.

16:30 - Welcome snacks and drinks

#### 17:00 - Talk 1

#### Why Context Matters in AI – Grounding LLMs

Aniket Mitra will explore how enterprise data captures the DNA of an organisation—the what, where, when, who, how, and ultimately the why behind decisions. He will explain why siloed data limits AI-driven decision-making, and how knowledge graphs provide a unified semantic layer that enables trustworthy, AI-powered control towers across the value chain.

Aniket Mitra – Aniket Mitra is the founder of Computationele Wetenschapsgroep (CW) which is a technology development SME that specialises in designing, building & scaling
Knowledge Graphs for enterprise customers across industry verticals like retail, manufacturing & logistics.

17:30 - Talk 2

Building Trustworthy AI in Regulated Sectors – Moonjelly, Knowledge Graphs, and the Context Layer
Surajeet Bhuinya will share how a dedicated context layer helps AI systems understand how organisations actually work. Using Moonjelly as a practical example, he will show how connecting people, documents, and processes into a living enterprise graph creates context unique to each industry and use case—leading to better decisions, clearer accountability, and real business impact.

Surajeet Bhuinya – Surajeet Bhuinya is the founder of DataPebbles, a leading Data & AI company in the Netherlands. With 25+ years of IT experience, he operates at the intersection of data, AI, and innovation, building scalable, practical AI systems that deliver measurable business impact.
He is also the founder of Moonjelly, a platform that connects people, documents, and processes into a living enterprise graph, enabling teams to understand context, see impact, and act with confidence.

18:00 - Talk 3

ReviewGraph – Using Knowledge Graphs to Predict Customer Satisfaction and Extract Hidden Insights
Lifeng Han will present ReviewGraph, a graph-based framework that transforms customer reviews into knowledge graphs to predict ratings and uncover hidden insights. He will compare this approach with traditional NLP methods and LLMs, highlighting how knowledge graphs offer strong performance with lower computational cost, greater interpretability, and promising integration with RAG systems.

Lifeng Han – Researcher, 4D Picture Project. Dr. Han has authored 50+ peer-reviewed publications and previously worked at the University of Manchester on Healthcare NLP. His PhD focused on machine translation (ADAPT Centre, DCU, Ireland).

18:30 - 20:00 - Networking, drinks and snack