You Deployed Your Machine Learning Model, What Could Possibly Go Wrong?
Training a highly accurate ML model is hard. Deploying it in production is even harder. Even if you manage to get it all together, things do not stop here. There is plenty that needs to be done after a model is deployed. In this presentation, we will talk about why is deployment not the last step in the ML model lifecycle, what all could possibly go wrong after model deployment and finally, how can that be prevented?
Surabhi is a Senior Machine Learning Engineer at Adobe building products for document intelligence. She is passionate about improving experiences using AI and ML techniques. Surabhi holds a master’s degree in Machine Learning from Columbia University and completed her undergraduate studies from IIT Guwahati, India. Apart from enterprise use cases, she has conducted research in the fields of healthcare and social good.