PhDs are usually more experienced and versed in algorithm development, research, and the theoretical side of ML. Practitioners with years of experience, on the other hand, focus more on using the tools and algorithms they know and are more application focused. Practitioners are the people you need for most projects as they have more experience with real-world problems and usually have more domain knowledge.
Theory vs Practice
Having more hands on experience can be a two edged sword. On one hand, having more experience means that you have a better and more in depth understanding of the problem, any potential solutions and the business applications of those solutions.
But as the old saying goes, if all you have is a hammer every problem looks like a nail. And this is precisely the problem with people who’ve worked on a single problem for years. They can be stuck on using the same old solutions to new problems. This limits your company’s potential to innovate.
On the other hand Ph.Ds come straight from university where they’ve had a much larger exposure to new ideas. Furthermore Ph.Ds have studied the theoretical side of the problem as well. Advances in theory can often times produce new and innovative solutions. This makes Ph.Ds more flexible and full of new ideas.
However, spending too much time learning the theory behind a given problem can make a person ignorant of the real world applications of their research. A good example of this is OpenAI’s “dangerous” fake news AI. The scientists at OpenAI developed an algorithm that can write as well as a human and just before publishing they realized that it can be used to generate fake news.
Let’s not mention the proclivity of Ph.Ds to waste all the resources at their disposal to get to the bottom of a problem which can rank up some serious costs to your business.
Should I get an experienced practitioner or a Ph.D data scientist for my team?
If you’re trying to build a cutting edge algorithm then PhDs are your people in most cases. For a well-balanced team, you need both. Practitioners can help you find and implement the solution to your problem while people with PhDs can provide more theoretical insights and help you improve your solution to surpass your competition in the long run.
Netlyt is an AI-only research and development company. We tackle non-trivial problems using the latest technologies and our own solutions. We’ve built systems to prevent corruption in the police force, track and predict air pollution, reduce natural gas distribution costs, detect faults in the electricity grids in Africa and others.