Machine learning at small organizations

Data Challenges

Small companies often face significant challenges in implementing machine learning (ML) due to perceived limitations in data and infrastructure. Kirsten Lum highlights that many small organizations underestimate their potential, believing they lack the necessary data or infrastructure to support ML initiatives 1. However, she argues that modern tools have simplified the process, allowing even single data scientists to make substantial impacts. The key lies in understanding that MLops at small companies doesn't require the complexity seen in larger firms. Instead, a simplified approach to deploying and monitoring models can be effective 2.

You just need to know MLops well enough to where you could deploy your own models, have a very simple batch pipeline that you know how to stand up in the technology that's available to you at your company.

--- Kirsten Lum

This approach enables small companies to integrate ML into their operations without overextending their resources.

   

Strategic Shift

Strategic adoption of machine learning in small organizations requires a shift in mindset and education. Kirsten Lum emphasizes the importance of measuring excellence by the impact of data science within specific verticals rather than just state-of-the-art performance 3. She notes that a strong A/B testing framework is crucial for demonstrating the value of data science projects, which can help earn trust and drive adoption in small organizations 4.

If we can point towards data science outputs as being driving impact, not just the accuracy of our models, then I think we'll see that adoption really start to grow within these smaller organizations.

--- Kirsten Lum

By focusing on delivering tangible results, small companies can better integrate ML into their business strategies.