Maximize Impact, Minimize Risk: Mastering ML in Production

We'll quickly define a "model in production." There's a myriad of definitions people use right now. The exact way we define it depends on several factors, including the size of the company, number of models, properties of the data, and so on.
While we're at it, we'll also answer some pressing questions such as: Is it ever OK to duct-tape the model deployment process? What about using shortcuts and opting for some manual work?
Practice makes perfect, or at least closer to perfection. We'll briefly present two case studies of solutions solving the same problem for two clients, both implemented with significant differences. We'll explain why. Both solutions required optimising search for results that translate into higher revenue. Yet, the companies are on opposite sides of the scale—one a large, mature retailer and the other a smaller, younger online booking business.
Cost must always be justifiable. Using these cases, we'll succinctly show how to fit a solution to match the context of the organization and utilise it effectively. Then, as we quickly go through the details of the models and infrastructures, we'll explain the reasoning behind the critical decisions and highlight the pros and cons of both approaches.