Six Best Practices for Bringing a Machine Learning Project into Production



Taking Machine Learning from the lab and into the real world is often the biggest challenge facing data scientists. In this talk, Lina Weichbrodt and Elena Stein from Taktile, a leading interactive decisioning platform for the financial services industry, will share six best practices they learned bringing dozens of machine learning models into production. As it turns out, the model is rarely the weakest link in making a project succeed. We will delve into go-live checklists, monitoring best practices, and what requirements should be defined in the business case. The talk will be most helpful for listeners with beginner or mid-level skills in productionizing data science products.


Lina Weichbrodt

Pragmatic Machine Learning Consulting, Strategy, Implementation

Lina has 10+ years of industry experience in developing scalable machine-learning models and bringing them into production. She currently works as a pragmatic machine-learning freelancer and consultant. She has helped clients in fintech, mobility, and travel to get value out of their AI projects. Come by and say hi, especially if you are also working on how to best use large language models in a conversational product.

Go To Speaker

Elena Stein

ML Engineer

Elena is a Machine Learning Engineer at Taktile, a leading interactive decisioning platform used by lenders and insurers worldwide. In her role, she helps customers to get the most out of the Taktile Platform and supports them in putting their models into production while using ML Ops best practices. She also develops models and implements business rules for projects in FinTech and InsureTech. Elena has previously worked as a Data Scientist for Klarna building credit risk models and as a Markets Analyst for Itaú BBA in London.

Go To Speaker