Aldís Elfarsdóttir

A fun fact about me is that I was born on the Stanford campus at the university hospital. My parents met here as graduate students; my mom was in medicine, and my dad was in electrical engineering. I’ve always had this romantic vision of Stanford because of their story. When I was 8 years old, we moved to Boston for my dad’s new job and I had a lovely childhood there, but returning to Stanford for my own graduate experience felt like a homecoming. 

I missed the greenery and the smell of eucalyptus and redwood trees, but I was also attracted to Stanford’s PhD program in Management Science and Engineering (MS&E) because it met the requirements I was looking for. I have always been interested in sustainability, ever since taking a high school class on the biology of alternative fuels and environmental sustainability. In college, I locked into environmental engineering, and I realized I needed higher education to secure the types of jobs I wanted and to pursue a more systems-level perspective. The Stanford program was attractive because it offered exactly that, blending an engineering component with a management component. 

Management science is wonderfully broad. It has many branch fields within the same department. I chose the energy and environment track. Basically, I learned the science of how to manage clean energy transition for governments, corporations, and communities around the world. I took classes along those lines, learned how to analyze economic policies, run cost-benefit analyses, utilize data science to make data-driven decisions at an organizational level, and I was also a teaching assistant for classes on those topics, as well as on managing organizations with data science.

The most transformational class I took here was Fundamentals of Data Science, taught by Professor Ramesh Johari. It teaches students to process data and generate insights from regression analysis and machine learning models, and to have a robust understanding of statistical concepts like making inferences from your sample to gain insights about the broader world and make generalizations with uncertainty claims. To me, that is one of the most valuable skills I have coming out of MS&E. I can apply to advanced positions in research and data science at a variety of organizations, from startups to giant corporations. Even with the current AI boom, you still have to know how to do data analysis.

My research is animated by the idea of “walking the talk.” When it comes to companies and climate change, it’s really important to me that the things they’re saying, like reducing carbon emissions, are actually happening. I recently won the best PhD presentation at the 2026 Sustainability Data Science Conference for my talk on corporate carbon credibility. I did that research using survey responses to detect if there are any credibility signals from the text data by applying natural language processing. I’m very aware of the concerns people have about AI’s environmental footprint, so I use it thoughtfully in my work by imposing limits on the number of prompts I issue and making sure I’m prompting for the highest payoff per prompt in terms of information gain and productivity. I treat it as auxiliary support and ultimately always rely on my own brainpower. After all, you need to know your work inside out if you’re putting your name on it as co-author or instructor.

One of the biggest takeaways from my findings is that specificity and consistency matter. Corporations will need to be really specific about how they’re calculating their footprint. They’re going to need to develop benchmarks for the industry to keep track of AI’s footprint, make sure it’s being used responsibly, and start reporting that data. That could be something I look into, but I also have this long-running interest in waste management. In college, I was an eco-warrior in my dorm, making sure that people were closing their windows in the winter and recycling and composting properly. I even did a whole study with sensors and surveys to reduce the energy waste from opening windows in the winter. I want to carry forward that spirit behind the next data analysis work I encounter, and hear from people about how they’re impacted by optimization recommendations like that.

I don’t have one specific dream job. I could see myself working with the United Nations Environment Program (UNEP) to follow up with companies and governments about methane leaks from four different types of sources: landfills, agriculture, energy and industry, and land management. As a post-grad I worked on a project at the Netherlands Institute for Space Research (SRON) developing a machine learning pipeline with human supervision to detect methane leaks around the world. Since I handed off that project, it’s been further developed to underpin a global system of methane alerts and responses (MARS) by the UN. I would love to circle back and apply my observational causal analysis skills from my data science experience to that kind of problem. That possibility feels within reach because Stanford’s network is really strong. I just attended a Stanford Alumni Alliance on Innovation for Global Impact event, and there were people there who had connections in the UN. Alternatively, the UNEP has a sustainable and circular textile initiative, and it would be cool to pursue research and development work along those lines. I would also be very happy working at a startup where you’re not limited to data pipelines and research but also get to see the impact you’re having on the broader organization and client outcomes. 

I love how many different ways Stanford has allowed me to explore my passions for sustainability and climate. While here, I even developed a side passion of crafting clothes from upcycled textile waste after taking a class called Data Visualization with Textiles taught by Quinn Dombrowski. The class introduced me to the Textile Makerspace on campus, where sewing machines, looms, fabric and yarn, and really helpful staff are available to help students design with textiles. My favorite creation is the dress I wore to present my PhD defense. I call it my PhD dress. It’s the first big upcycling project I undertook, and I made it using these horizontal scraps that I sewed together vertically to make a dress. I also made quilted pants using old pillowcases from my aunt’s house as the inner lining. It’s really fun, but there’s a learning curve; the first dress I made using actual PhD data shrunk in the wash and I couldn’t actually wear it! In the end, my romantic vision of Stanford held up. I fell in love with so many things here – from the campus to the community, and from fascinating research topics to expressive art forms.