Explore the journey from robotics to recommender systems in this insightful podcast episode featuring Miguel Fierro, Principal Data Science Manager at Microsoft. Delve into the limitations of applying machine learning in robotics, the integration of computer vision and AI in sports analytics, and the evolution of recommender systems. Learn about the importance of choosing simpler solutions over complex ML models, the role of embeddings and feature stores in modern AI applications, and strategies for demonstrating ROI to leadership. Gain valuable insights on high-impact AI investments and the potential of Large Language Models in recommender systems. Perfect for data scientists, AI enthusiasts, and business leaders looking to understand the practical applications and challenges of AI across various domains.
From Robotics to Recommender Systems - MLOps Podcast #240
Overview
Syllabus
[] Miguel's preferred coffee
[] Takeaways
[] Robotics
[] Simpler solutions over ML
[] Robotics and Computer Vision
[] Basketball object detection
[22:43 - ] Zilliz Ad
[] Mr. Recommenders and Recommender systems' common patterns
[] Embeddings and Feature Stores
[] Experiment ROI for leadership
[] Hi ROI investments
[] LLMs in Recommender Systems
[] Wrap up
Taught by
MLOps.community
Related Courses
-
Recommender Systems with Machine Learning
-
Recommender Systems: An Applied Approach using Deep Learning
-
RecSys at Spotify - Building and Evaluating Large-Scale Recommender Systems
-
Unpacking Three Types of Feature Stores - MLOps Podcast
-
Who's MLOps for Anyway? - Navigating AI Initiatives and ROI
-
Collaboration and Strategy in Data Science and AI - MLOps Podcast
