My Journey from Microsoft to the Cleantech Industry
[This post first appeared on the EnergySavvy blog---Eds.]
In May 2008, I left my corporate software job in search of my next challenge. I had spent 10 years developing software in various product roles mostly at Microsoft.
When I left, I had heard of the hyped-up cleantech space and said to myself, “I don’t know anything about it, but if there’s a way to apply software skills to societal goals in a way that can be valuable, count me in.” Then I learned about energy efficiency.
Energy Efficiency Software Challenges Abound
After co-founding an energy efficiency software company, EnergySavvy, in 2008, I’ve been surrounded by interesting software challenges for the last two years. This space has barely had a Web 1.0 (let alone a Web 2.0) and is entering a great modernization phase.
My hope is that this will inspire at least one other software person to think about applying their software skills (h@cker Skillz?) to energy efficiency. So here’s a round up of interesting software challenges. To keep it simple, we’re going to focus in on three areas of high activity: energy monitoring and analytics, energy data, and building analysis.
Energy Monitoring and Analytics
Lots of companies are focused on energy monitoring from real-time smart meter and automation companies like Silver Spring Networks to monthly reporting like OPOWER which compares usage to one’s neighbors. The heart of the idea is if you give more awareness on one’s energy use, you can trigger action.
The challenge is what do you do with energy usage data beside just graph it? Clearly, we as a software community can do better. What about:
* Doing social comparisons aside from just one’s neighbors? How about usage of homes of similar specs such as age and square footage? How about relative to one’s friends or family? In that case you have to normalize for energy prices, energy intensity (energy use per square foot) and wire it up to some kind of social network.
* How about analyzing usage data in aggregate to give users specific goals based on observed achievable savings based on some peer group?
* And what of user interface? Few homeowners are equipped to deal with smart meters, and homeowners often express frustration with the current smart meter tools.
* Most of these monitoring devices completely skip natural gas (let alone oil) usage, so that seems relatively greenfield albeit hard.
If you’re a network protocol hacker, you should check out ZigBee, which is often the protocol of choice for energy monitoring objects in the house. Also, check out OpenEnergyMonitor.
Residential Energy Data
One of the issues of doing residential energy efficiency at scale is identifying where the savings lie. There are tons of data sources out there to map together to help with this:
* Take a look at the Energy Information Administration’s (EIA) website. If you’re a data hacker and interested in cleantech, it’s a feast for sore eyes. Tons of data. What if we mashed the EIA data together with real estate data or GIS data, especially with thermographic mapping?
* Another area is making basic savings predictions based on energy intensity—the measure of energy use per square foot of a home. With basic bill data, square footage and age of home data from the MLS, you could probably get something really interesting going.
* Still another area is statistical analysis. Utilities have huge energy efficiency targets to reach (see here). One of the key challenges is figuring out what is cost effective and what’s not. For example, if they subsidize a CFL bulb, they need to measure kilowatt-hours saved, how much is significant to the program vs. what would have happened anyway via regular market adoption, what’s the actual savings per dollar, etc. This gets complicated quickly and goes bezerko when carbon legislation kicks in.
Building Analysis and Interoperability
One of the hassles of the industry is predicting the energy use of a home based on an in-home energy audit. There are a ton of tools for this of varying accuracy and complexity. If you’re interested at all in this space you need to know about the HOME STAR Bill. HOME STAR will offer a $3,000 tax credit if a home is made 20 percent more efficient. So, the question is, regarding 20 percent, who says and with what tool?
A few challenges and opportunities:
* Obviously, there can be algorithmic improvements here—you’ll need to know building science, which fortunately isn’t rocket science. The main introductory textbook for the field is Residential Energy.
* There are big workflow and modeling issues with home energy audits. Recurve is a software company working on this, and I’m sure they are hiring.
* Interoperability is a huge issue here. There are literally hundreds or thousands of different energy programs that industry participants need to work across. Of note is financing, which is growing quickly in this space in both secured form (see PaceNow.org) and unsecured. The financing systems (themselves nascent and a software opportunity) and the audit sources need to communicate, so we need interop. Some folks have begun working on this, but there’s a long way to go.
Wrapping Up and a Shameless plug
Hopefully this gives you a sense of some of the cool challenges at the intersection of energy efficiency and software. And this is just a small sampling of some of the opportunities.
Our own company, EnergySavvy, is ramping up quickly, and we’re hiring. I didn’t delve much into what we do, but suffice it to say, we have big technical challenges ahead. Currently, we use Python, Django, Postgres, PostGIS to weave together energy modeling algorithms, energy data, rebate info, utility and municipality service area mapping and workflow. And that’s just the beginning—we have big plans ahead. Think of what you see at EnergySavvy.com as merely a prototype from a technical perspective.
We also started a LinkedIn group for developers working in energy efficiency, or developers who want to make the transition into cleantech but aren’t sure where to start.