Expert Interview (Part 1): Tobi Bosede on What It Takes to Be a Machine Learning Engineer
At the Strata Data Conference in New York City in the fall, Paige Roberts of Syncsort had a chance to sit down with Tobi Bosede, Sr Machine Learning Engineer, shortly after her presentation. In the first of this three-part blog series, Bosede explains what goes into being a Machine Learning Engineer as well as some of the projects she is currently involved with.
Roberts: So, tell everyone a little about yourself.
Bosede: My name is Tobi Bosede. I got my graduate degree from Johns Hopkins and my presentation was on the research I did for a graduate thesis. I am a Machine Learning Engineer so I look at all sorts of data relating to finance, not necessarily relating to trades, but that includes staying up to date with current tools and technologies such as python libraries or using things like Ansible. Implicitly even though by title I’m a Machine Learning Engineer, I’m also a Data Engineer.
You have to be. Do you do a lot of your work on the cloud?
Yeah. Some of it is local on my machine but, yeah. A good amount is on the cloud, especially things that we productionize or even when we are doing a demo, and we are all trying to work together on something we’ll put it on a AWS for instance. There are so many technologies and it’s fast changing and fast moving so a huge part aside from research is just staying up-to-date by going to conferences, talking to peers at various companies, and reading about current technologies because it’s always changing.
Keeping up because it always changes. Yeah, I get that.
In addition to being a Machine Learning Engineer, I’ve been involved in NumFOCUS. I have done some extracurricular things with them. Have you heard of NumFOCUS?
No, I haven’t, tell me about it.
Well, it’s a non-profit. It was kind of spun off from Continuum, to support Python libraries like NumPy, SciPy, Pandas. Those are all NumFOCUS projects. So, basically the idea is to fund open source projects that maybe would have difficulty staying up to date and being maintained. It’s actually pretty time consuming to do all that open source work. Wes McKinney created Pandas but he also has a day job. The larger the number of consumers or users of your tool and your library, the more intensive maintenance and upkeep on it is.
You have to have a fairly good community to keep it up.
Make sure to check out part 2 where Bosede will explain predicting trade volumes and the correlation between volume and volatility.
For more on machine learning, read our white paper: Why Data Quality Is Essential for AI and Machine Learning Success