Sramana Mitra: Wearables is a very interesting industry that is going to be entirely design-driven. What about on the IT side whether it’s a healthcare IT or software-driven solution? What have you invested in that is interesting?

Dheeraj Pandey: We’ve done a late-stage investment in a Healthcare analytics company. They’re trying to surround Epic, which is, kind of, the mainframe of healthcare. It’s a very interesting play because they’re not going directly after the “mainframe”.

How do you really take healthcare to the cloud? The overall delight starts to come from the end users back to providers. It becomes a very useful way to sell your technology to big hospitals. You start to see that you’re engaging with your patients. Then it also is helpful for insurance companies because they start seeing how it could reduce their liabilities if you are in the space of population health and having patients come back on time as opposed to being reactive. How do you make them proactive?

I love this space because it’s not going head-on after something that’s very stodgy and extremely bureaucratic in the way decisions are made in healthcare given how much regulation there is. Over time, they probably will. They will get to the main screen of the doctor when they log on. Doing this outside in has been extremely innovative. I’ve been a big fan of such a company as well.

Sramana Mitra: You come from the data world. How do you see the trends and opportunities in data? If you were starting a company today in the data space, are there problems that could be opportunities for new entrepreneurs?

Dheeraj Pandey: You have to do this very inexpensively because people do a ton of experiments in data. Converting data to information still requires a lot of domain knowledge. What we are doing at DevRev is taking a ton of this data and software engineering. How do you create canonical models so that people don’t have to create data warehouses and do analytics on top of that? It’s still a pretty expensive thing to crunch data into information.

Finally, visualizing is a pretty hard problem. The biggest challenge is data engineering. Many data scientists are not good at engineering. They’re good at science but not good at engineering. That includes collection all the way to masking, transformation, and loading it to massive databases. Then run some compute on it because you might have to run some logic before it gets ready. All that is still a pretty complicated thing. A couple of companies have done it well. It’s still not available to every industry out there. We probably have made something for the tech world but need to really make it pervasive and bring it to design.

I’m a big fan of the New York Times’ visualization. It saved the company. Most of these papers were going down the drain. You could say it’s because of their subscription model and they built their own ad network. At the core, I think New York Times is a design company. It’s not just known for its font but the culture of design has been a big force of innovation for them. Developers are visualizing with new tools. Beyond developers, it’s still a big challenge.