Speaking

Keynotes and invited talks on AI systems, efficient inference, agentic AI, and hardware, plus recorded talks and lightboard explainers from the IBM Technology channel.

Efficient LLM Inference: From Cloud to the Edge — Hardware-Software Co-Design in Practice
Keynote

Efficient LLM Inference: From Cloud to the Edge — Hardware-Software Co-Design in Practice

ICDTA-Global 2026

Inference, not training, now drives the cost of running LLMs in production, and long context, multimodal inputs, and agentic workloads make it harder across cloud and edge. This keynote works through hardware-software co-design: quantization, efficient attention, and KV-cache optimization pay off only when they map well to the hardware, with examples from IBM Granite, the llm-d distributed inference framework, the Spyre accelerator, and research architectures such as NorthPole and analog in-memory computing.

Vaughn College, New York
Oct 2026
View the program

Efficient LLM Inference: From Cloud to the Edge — Hardware-Software Co-Design in Practice

ICDTA-Global 2026

Vaughn CollegeOct 2026

The Future of Computing: IBM Research and the Rise of Generative Systems

IBM Technology in Action DACH

EhningenJul 2026

Efficient LLM Inference: From Cloud to the Edge — Hardware-Software Co-Design in Practice

International Summer School on Generative AI 2026

RomeJun 2026

Agentic AI: From Models to Autonomous Intelligence

Orange Morocco Agentic AI Day — Trust the Future!

CasablancaMar 2026

Agentic AI: From Creation to Collaboration

Women in AI Morocco / INPT Workshop

INPTDec 2025

The Next Wave: Reinventing Intelligence and Compute Architecture

Women in AI Morocco Summit 2025

TechnoparkDec 2025

Agentic AI Workshop — Full Auditorium

Women in AI Morocco / INPT

INPTDec 2025

The Future of AI — An IBM Research Perspective

IBM TechXChange

Johannesburg & Cape TownAug 2025

Revolutionizing Enterprise AI: The Power and Promise of Foundation Models

Women in Research Webinar Series (QUWA) — University of Sharjah

SharjahApr 2025

Revolutionizing Enterprise AI

IEEE Services Conference

ShenzhenJul 2024

Scaling Foundation Models for Enterprise

MoroccoAI / Al Akhawayn University

Ifrane2023

Foundation Models at Scale

AI Seminar Series — Alfaisal University

RiyadhFeb 2023

Women in Computing: Breaking Boundaries

ArabWIC Conference

RabatMar 2019

AI for Business: A Unique Set of Challenges

Women in Data Science @ Stanford / KACST

Riyadh2018

Watch Talks & Videos

Selected recordings from conferences, events, and explainer videos.

Lightboard2026

AI Model Debugging: Find the First Divergence Across AI Accelerators

When a model gives a wrong answer on a new accelerator, the fault can sit in any of its dozens of layers. Run the same input on a trusted reference, compare the output of every layer, and start at the first one that diverges.

IBM TechnologyYouTube
Lightboard2026

Why AI Models Run Faster on Some Chips: The KV Cache Effect

A lightboard explainer on the IBM Technology channel on the KV cache, the model's notes on what it has read, and how that state moves differently across a GPU, a TPU, and IBM Spyre.

IBM TechnologyYouTube
Lightboard2026

The AI Memory Wall: Why LLMs and AI Accelerators Hit a Bottleneck

A lightboard explainer on the IBM Technology channel on why memory bandwidth, not compute, is the real limit for LLM inference, and the techniques that work around it.

IBM TechnologyYouTube
Keynote2020

Al Akhawayn University — Alumni Journey

A personal reflection on the journey from Al Akhawayn University in Morocco to IBM Research, sharing how the university's liberal arts education and international environment shaped a career in AI and computer science.

Al Akhawayn UniversityYouTube
Keynote2021

Powering the Future of AI through Specialized Hardware

Keynote at MoroccoAI Annual Conference discussing how specialized AI hardware accelerators are essential for sustainable and efficient AI, covering analog in-memory computing and hardware-software co-design strategies.

MoroccoAI Annual ConferenceYouTube
Keynote2021

Accelerating, Optimizing, and Automating AI across the Stack

A comprehensive keynote on the challenges of deploying complex AI models efficiently, covering optimization techniques from hardware to software, and automated approaches to neural architecture search.

ECOLE DES PONTSYouTube
Keynote2022

Platform for Next Generation Analog AI Hardware Acceleration

Presentation at the tinyML On Device Learning Forum on building platforms for next-generation analog AI hardware, enabling efficient on-device inference through novel computing paradigms.

tinyML ForumYouTube
Keynote2021

Women in Services Computing — Award & Keynote

Award acceptance speech and presentation at the IEEE International Symposium on Women in Services Computing, highlighting contributions to AI research and inspiring the next generation of women in technology.

IEEE WISC 2021YouTube
24+Keynotes
10+Countries
20+Invited Talks
5Continents