In Episode 18 of Season 3, of Driven By Data: The Podcast, Kyle Winterbottom is joined by Sarah Schlobohm, Head of AI at Kubrick Group, where they discuss how not to do Machine Learning and why organisations are getting it wrong, which includes;
- The frustrations as a hiring manager in the ML space
- The most common mistakes organisations make when implementing AI
- The importance of engineering in ML
- Why so many ML projects don’t make it into production
- Why you need to think about the whole lifecycle and ecosystem
- The different challenges from the technical side to the business side
- Why you don’t want to be a solution looking for a problem
- Why you should fail and do so fast
- The importance of knowing when to give up
- Balancing the business strategy and the hype of AI
- The signs that you’re not ready for AI/ML
- Why you should do some analysis to identify relevant use cases
- Why you should have the technical skills in the room before the ML decision is made
- The future of ML tooling
- Why the cost of ML is drastically underestimated and the impact that has on ROI
- Why the amount of time and effort that should go into Data Governance is underestimated
- The role of MLOps
- Why there’s no substitute for domain expertise
- The pros and cons of different reporting structures
- AI Ethics and the examples of where things have gone wrong
- The practical steps that organisations can take to improve Diversity & Inclusion
- Why diversity is a cultural challenge
- Why women shouldn’t be pushed into non-technical roles
