MLOps community #3 (in-person | en personne)
Details
Access:
- The entrance is located on the side of Marconi Street. There will be someone at the door to assist you, or there will be signs directing you.
- Additionally, we kindly request that you provide a proof of registration in order to prioritize access those who are registered.
Join us for an evening of cutting-edge discussions and valuable insights at our last meetup before the summer break! We are thrilled to announce that this event will take place in the office of Sama in the Mile-Ex.
Get ready to immerse yourself in a series of captivating talks that delve into the latest advancements in MLOps. Don't miss the opportunity to gain valuable knowledge and connect with experts in the field.
Agenda
5:30 pm - Doors open
6:00 pm - Introduction
6:10 pm - MLOps in Computer Vision with Humans in the Loop by Frederic Ratle, Akshay Pardhanani & Marc-Antoine Vézina (Sama)
6:50 pm - Break and networking (drinks and pizzas will be provided)
7:10 pm - MLOps for Online Inference by Olivier Labrèche (Shopify)
7:50 pm - Final notes, break, and networking
Talks
MLOps in Computer Vision with Humans in the Loop (Sama)
In this talk, gain insights into the problems Sama is solving in the computer vision and AI-assisted annotation spheres. Explore the challenges in data wrangling and the tooling that enabled the teams to work around the issues. Furthermore, discover how Sama leverages human-in-the-loop to gain feedback about the model's performances and the infrastructure surrounding them.
About Frederic: He is currently Head of AI at Sama, leading machine learning research and development. He has nearly 15 years of industry experience in machine learning, computer vision, natural language processing, and speech recognition. Before Sama, Frédéric held research and management positions at Nuance Communications, where he improved speech and NLP systems for Healthcare and Automotive customers, and at AutomatAI, he built conversational product recommendation systems. He holds a B.Eng. and M.Sc. from the Ecole Polytechnique de Montréal (Canada), and a PhD from the University of Lausanne (Switzerland).
About Akshay: He currently serves as the tech lead for MLOps at Sama. He has eight years of experience in Big Data and MLOps with organizations such as Accenture, SkipTheDishes, and Clearco. His experience and skills lie in building and scaling infrastructure to operationalize ML Models, including but not limited to, Computer Vision models, regression models, and classification models.
About Marc-Antoine: He has four years of experience working on ML-oriented projects. Previously working at Intact as an AI Engineer in operationalizing ML Models and building feature engineering code, he now works as a MlOps Software Engineer at Sama where he helps streamline model deployment, data wrangling and CI/CD. He is passionate about data engineering, automation and X-as-Code.
MLOps for Online Inference (Shopify)
In this talk, we focus on online inference, a crucial portion of MLOps where models are deployed and served in order to respond to requests on-the-fly. We present the various components involved in online inference, discussing the platforms and tools that Shopify, a world leader in eCommerce, has iterated over the last few years. We will end with giving a quick overview of the future direction of our machine learning platform.
About Olivier: He started his career as a space telecommunications engineer. His interest in AI developed over time with the release of increasingly powerful ML applications and the growing importance of the field in Montreal. His ML journey across different companies such as Element AI, BNP Paribas and Shopify brought him from model building to MLOps as he realized that many challenges came from the lack of established processes and tooling. He is a multidisciplinary engineer currently focusing on back-end systems at Shopify.
Let's make this meetup an engaging and informative experience! Don't miss out on this opportunity to learn from industry experts and expand your professional network.
