Lindsay Baker
Lindsay Baker

Building systems are getting smart, and their ability to collect data is helping designers and developers better understand the occupant experience—in theory. Still a relatively new resource, the resulting data can be difficult for the average building operator to parse. At the International Well Building Institute's symposium held on Oct. 20 in New Orleans, we talked with Lindsay Baker, vice president of business development at the Silicon Valley startup Building Robotics, which makes the intelligent HVAC occupant control system Comfy, about the collection of occupancy data and its ability to shape a space.

What is the state of innovation for startups in the intelligent building–systems space?
It’s an exciting time to be in the automation market for physical space. There’s always been a need to understand what’s happening in a space, but we never imagined that we would be able to gather information in a way that is this granular. The infusion of technology into the built environment has the potential not just for convenience but for improving the quality of life and decreasing buildings’ environmental footprint in a profound way. It’s not all about reducing usage; it’s about users getting what they need and nothing more. Also, issues like inadequate ventilation and natural light disproportionately affect people who aren’t empowered to say that they want a better environment, like kids who live in housing projects in inner cities. The ability for us to start sensing their environment helps us build a data-driven argument for better buildings.

One of the most interesting parts of our work is the intersection between technology culture and real estate culture. Technology has impacted the consumer world very fast, but there are a lot of steps in getting something into a commercial building. These big real estate companies own lots of buildings and it’s hard for them to take one technology and implement it in all of them. One of the biggest challenges we have as a company is to be a startup and achieve that level of impact within an industry that has typically been one of the slowest to adopt new technologies.

During your talk at the Well Building Institute's symposium, you said that there is “a lot to be managed” in the creation of healthy spaces. Can you explain that statement?
There’s a risk of collecting too much data in buildings and not knowing how to digest it. We have to do a better job of curating the right level of information. We’re not quite there yet because we’re still getting used to the fact that we can provide this information. Space utilization is a great example. The lighting-controls companies have realized that installing intelligent lighting gives them all this data about how people use space. They know they have that data, they can show you the data, but how does that data become a necessary part of the feedback loop in how we create buildings? That’s still a challenge, and why we at Building Robotics have decided to provide a level of automation. Comfy is very bounded; building managers choose how warm and cool a space can be and occupants are allowed to adjust by about 8 degrees [F] within that zone. It’s a lot more than the 2 degrees [F] that is typical. Our hope is to be a connection between people and the building.

What role does machine-learning play?
A building manager doesn’t have time to cater each room to someone’s preferences. But machine learning can make sure that buildings respond to these dynamic preferences over time. People are getting to understand and expect a little machine-learning in their lives. For example, when you type a search entry term into Google, it comes up with a few guesses as to what you might be searching for. So when people walk into their office, there will be a growing expectation that the office should be adapting to them because they’ve told it before what temperatures and lighting levels they like. That’s where machine-learning gets exciting—it can pick up on those trends without people feeling like they have to constantly provide feedback to the system.

This interview has been edited and condensed.