Navigating the Maze of Production Machine Learning
Turning Complex Theories Into Practical Real World Solutions - Miguel Otero Pedrido
Successful machine learning requires engineering robust systems that thrive on real world data. Building the complete surrounding infrastructure ensures that AI applications deliver continuous value outside the laboratory. Miguel Otero Pedrido knows exactly how to bridge the gap between complex theory and practical deployment.
In our upcoming episode we will explore how to build reliable AI infrastructure. From his journey of studying quantum field theory to deploying fashion recommender engines and complex agentic workflows. We will also discuss the essential engineering practices needed to take an AI project from an initial concept to a fully operational production system.
He’ll cover:
Transitioning from theoretical physics to applied machine learning
Why good models fail without proper system engineering
Building recommender engines and complex agentic workflows
Moving beyond hype to construct reliable AI applications
Core engineering skills for real world machine learning production
Lessons learned from founding The Neural Maze
About the Speaker:
Miguel Otero Pedrido is an ML AI Engineer who learns and teaches by building. He went from studying spacetime curvature and quantum field theory to creating real world ML systems ranging from recommender engines in fashion to complex agentic workflows for insurance companies. Along the way he realized good models are not enough since engineers must build the whole system to make AI actually work. Now he shares what he learns through The Neural Maze as his way of helping others build AI that goes beyond the lab. He is always learning always building and always up for a good chat about math systems and how to make AI useful.
DataTalks.Club is the place to talk about data. Join our Slack community!