In this talk, we introduce the concept of AIOps referring to using AI and data-driven tooling to provision, manage and scale distributed IT infra. We particularly focus on how AIOps can be leveraged to help train and deploy machine learning models and pipelines at scale. From provisioning the right amount and type of GPUs and CPUs, to selecting the right cluster and cloud provider, to understanding the relationship between quality of experience (QoE) metrics like model precision and serving latency from one hand, and quality of service (QoS) metrics like processing speed, memory size and memory bandwidth from the other, we help you think through the right questions to consider when selecting, fine-tuning and deploying the ML models powering your business AI strategy.
Speaker
Ghida Ibrahim
Chief Architect, Head of Data @Sector Alarm Group, Ex-Facebook/Meta
Since April 2023, Ghida has been the Chief Technology Architect and head of Data at Sector Alarm, one of Europe’s top providers of smart home solutions and a KKR portfolio company.
Prior to that, she spent 5+ years as a technical lead at Meta/Facebook building AI tools and systems to help scale and optimize Meta’s edge computing infrastructure, used to serve billions of people across Meta family of apps.
Ghida’s experience also includes working for major European Telcos in roles at the intersection of distributed computing and advanced analytics. She holds a PhD and Masters in computer Engineering from Institut Polytechnique de Paris.
Outside of work, Ghida is an expert technical advisor on new technology trends for the World Economic Forum, the French Secretary of Investment, among others. she also occasionally lectures at university and speaks at conferences
Date
Tuesday Apr 9 / 10:35AM BST ( 50 minutes )
Location
Churchill (Ground Fl.)
Slides
Slides are not available
