Speaker
Abstract
Taking AI from lab to business is notoriously difficult. It is not just about picking which model flavor of the day to use. More important is making every step of the process reliable and productive. From training, experimentation, deployment, validation and everything in between, there are lots of moving pieces.
This talk will provide a high level overview of LinkedIn’s AI ecosystem, and then zoom in on the data platform underneath it: an open source database called Venice which we’ve been running in production for 7 years.
Building a data platform specifically tailored for AI requires some careful considerations. Among other things, it must support rapid experimentation, high throughput ingestion, and low latency queries for online inference applications.
You will come out of this session with an understanding of these various challenges, what we did to solve them, and how we pivoted along the way to keep up with changing workloads and requirements.
Topics
QCon London 2024 is a three day conference for senior software engineers, architects and team leads. An international program committee of working engineers selects every session. Patterns and practices, not products and pitches.
Part of the track
Architecture for the Age of AI Hosted by Fabiane Nardon Data Expert, Java Champion & Data Platform Director @totvsFrom the same track
Tuesday 9 April
10:35 Churchill (Ground Fl.) Session When AIOps Meets MLOps: What Does It Take To Deploy ML Models at Scale Ghida Ibrahim Chief Architect, Head of Data @Sector Alarm Group, Ex-Facebook/Meta 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. 11:45 Fleming (3rd Fl.) Session AI/ML Mind Your Language Models: An Approach to Architecting Intelligent Systems Nischal HP Vice President of Data Science @Scoutbee, Decade of Experience Building Enterprise AI As large language models (LLMs) emerge from the realm of proof-of-concept (POC) and into mainstream production, the demand for effective architectural strategies intensifies. 13:35 Rutherford (4th Fl.) Event Connecting the Dots: Applying Generative AI (Limited Space - Registration Required) Details coming soon. 14:45 Windsor (5th Fl.) Session Flawed ML Security: Mitigating Security Vulnerabilities in Data & Machine Learning Infrastructure with MLSecOps Adrian Gonzalez-Martin Senior MLOps Engineer, Previously Leader of the MLServer Project @Seldon The operation and maintenance of large scale production machine learning systems has uncovered new challenges which require fundamentally different approaches to that of traditional software. 15:55 Fleming (3rd Fl.) Session Large Language Models for Code: Exploring the Landscape, Opportunities, and Challenges Loubna Ben Allal Machine Learning Engineer @Hugging Face In the rapidly evolving landscape of software development, Large Language Models (LLMs) for code have emerged as a groundbreaking tool for code completion, synthesis and analysis. 17:05 Whittle (3rd Fl.) Session AI/ML Lessons Learned From Building LinkedIn’s AI Data Platform Felix GV Principal Staff Engineer @LinkedIn Taking AI from lab to business is notoriously difficult. It is not just about picking which model flavor of the day to use. More important is making every step of the process reliable and productive.