Each quarter, Chicago’s top digital leaders take a communal lunch break to meet, mingle and hear from Built In Chicago’s featured speakers. Yesterday, the group of 100 settled in at the ever-swanky RPM to hear from Scott Nicholson, Chief Data Scientist at Accretive Health, and Rayid Ghani, former Chief Scientist for Obama for America.
[ibimage==21786==Original==none==self==ibimage_align-center]Data, it seems, is becoming more important by the day. Proper handling allows organizations and businesses the opportunity to learn about, understand and even influence their audiences. Data is not without its problems, though – there’s so much of it that only a relatively few people are truly equipped to manipulate it, and there are frequently concerns about the privacy of those whose data is being analyzed.
[ibimage==21787==Original==none==self==ibimage_align-center]In the hot seat, Nicholson and Ghani took questions from Matt Moog, Founder & CEO of Viewpoints, and Founder of Built In Chicago, as well as guests. Here are some of the most compelling questions:
How do you view your field?
SN: For me, data science is less about the technical side and more about this end-to-end decision making process, starting with the questions...and then doing whatever you have to do to actually have an impact.
RG: It’s about, do you have a set of people who will allow you to access data, be self-sufficient, and do principled analysis that allows you to make meaningful decisions?
Which industries have the most future potential in terms of data?
RG: The way I view potential is two dimensions: one is how interesting, dynamic, detailed the data is...and the second is what’s your ability to take action on that and on what level? If you look at the work that’s going on in the online learning community...in addition to this data, you have channels where you can interact with individuals at their pace and really adapt, so I think that’s a really exciting space.
SN: Given that I jumped from consumer Internet/social media [LinkedIn] over to healthcare, I clearly think that healthcare is one of the most exciting industries right now. You have an interesting, pragmatic and almost boring problem to solve of having all this data out there, we need to make sense of it, we need to get people on board with why it’s useful, increase transparency. These things are not about deep data mining and machine learning; it’s really about, what are the problems that we’re trying to solve in this industry?
[ibimage==21788==Original==none==self==ibimage_align-center]If it’s all about asking the right questions, how does the data scientist work with the management team to come up with the right ones?
SN: In my experience, it’s harder to find people who know what the right questions are. It starts with hiring. I want to make sure I’m getting the right people on board. The other part is, I want to make sure to give them the flexibility to actually exercise creativity. Allowing people a venue for creativity is where, then, the important projects come out.
RG: If you hire the right people and put them in an environment where they’re around other people they can look up to and get a better sense of the business, smart people will figure out the questions and the best way to do that is lunch, like this right here. You put smart people together in the right context, they will figure out things.
How do you hire a data scientist?
RG: You want to find people who have a good, solid background in data and analytics but who also have common sense and the intersection is very hard to find. What do you choose? If you have to pick people who are common sense smart or analytics smart, which one would you go heavy on? My perspective is I’d rather hire somebody out of school for what they can learn in school, which is more the technical and data, and I can expose them to other people around me and they’ll pick up the common sense.
SN: Probably some of the best data scientists out there aren’t going to have their title on LinkedIn as “data scientist.” It’s going to be “software engineer.” For me, given that my goal is to build products, I want to make sure that it’s not people who ask questions, it’s people that can build something so actually I’m kind of biased toward engineers who have experience with data or are very much passionate about developing their career in that direction.
What advice would you give to growth-stage startups who are starting to get a lot of data but don’t have the infrastructure in place to handle it?
SN: The infrastructure piece is probably the hardest because it’s the most in-demand skill in this world of data engineering. There are tons of people in grad school who want to play with data and just don’t have an opportunity to do it. It could be a research partnership...that plus internships I think is a great way to get people who are excited about the work you’re doing but you just kind of leverage them. It’s not like you’re bringing on a full-time employee at who knows how much money.