June Meetup: High-Performance Data Meets Modern AI

Details

This international group, headquartered in Madrid, is dedicated to promoting the q programming language, the kdb+ database, and their ecosystem. All talks are technical and conducted in English, with a focus on creating a hybrid experience (in-person + streaming). If you like elegant code and astonishing speed, this is your place.

Agenda:

18:20 Registration and Welcome!
18:30 Umaima Bousdar Ahmed (BBVA) Leveraging Neural Networks in KDB Insights for Financial Applications
19:20 Jesús López (Habla Computing) Notes from Building with kdb-x
20:00 Networking

Abstract:

Umaima Bousdar Ahmed (BBVA)This talk presents the development of a neural network implemented natively in q/KDB Insights to estimate execution probability in a financial context. Its performance is benchmarked against a range of classical statistical and machine learning models built within the same environment. The session also outlines a real-time RFQ simulation use case to illustrate practical deployment considerations.

Jesús López (Habla Computing) kdb-x keeps the performance that made kdb+/q a go-to for high-throughput data systems, then gives it a modern shape: a module system, first-class Python interop, and new AI modules, alongside longer-standing pieces like dashboards.
This talk is a practical look at what that combination feels like when you build with it, drawn from real projects. A central thread is exposing a kdb-x system as an MCP server and wiring it into Solace Agent Mesh, so high-performance, real-time data becomes an active participant in agent-based architectures rather than a store the agents query from the sidelines.

Biography:

Umaima Bousdar is a Quantitative Analyst specializing in machine learning applied to financial markets, based in London. She is currently part of the Advanced Analytics & Algorithmic Trading team at BBVA (CIB, Global Markets), where she serves as a Senior Data Scientist within the Credit Algo Team, with strong experie