Earth Hacks: Tell us a bit about yourself (i.e., where you’re from, your background, what you’re in school for, etc.)!
Prajna Soni: Currently I am a Master’s student at MIT pursuing the Technology and Policy Program with a dual degree in EECS with a focus on responsible and ethical AI. Prior to pursuing my Master’s degree, I studied Mechanical Engineering with a minor in Computer Science at NYU Abu Dhabi; where I completed a year-long research fellowship exploring the global data divide and how technology shapes society, specifically looking at bias propagation in recommender systems.
EH: How did you get involved with Earth Hacks?
PS: I found out about Earth Hacks via the GREEN Program alumni community. Sanjana was a GREEN Program alum and I heard about her after attending the program a few years ago. At that time, I knew I wanted to do something computational and wanted to explore how it is currently being leveraged in environmental science and applications and so I looked into Earth Hacks and upcoming events! I was exploring satellite data for my senior capstone project as a means of monitoring and contributing to preservation efforts for mangrove forests and so the Urban Heat Hackathon with ECOSTRESS seemed like a great fit!
EH: What was your experience like at the Urban Heat Island Hackathon? Can you tell us about the project that you were involved with?
PS: The Hackathon was very informative and interesting! I learned a lot about the efforts at NASA that scientists like Glynn Hulley lead, specifically NASA ECOSTRESS. MY team’s project focused on mapping the urban heat island effect of Mumbai using ECOSTRESS data to aid in informing policy. We also wrote a MATLAB script to increase the resolution of the mapped heat islands from 70m as provided by ECOSTRESS imagery to 30m as provided by Landsat imagery using random forests.
EH: What drew you to be interested in the Tech & Policy space and what are your goals after grad school?
PS: After graduating from my undergrad, I did a one-year research fellowship exploring the intersection of technology and social science — specifically how recommender systems have the potential to propagate bias within user groups. I was interested in understanding this intersection further and came across the Institute of Data, Systems and Society at MIT. Their work really resonated with the kind of work I wanted to be doing and so I applied! I’m hoping to continue to work at this intersection, initially in more technical roles but ideally always on problems that have a wider impact and in a capacity to try and contribute meaningfully. It’s still an open question but I’ve had the opportunity to learn from and work with very interesting people on interesting projects at and beyond MIT so I’m excited for the future!
EH: What role can Artificial Intelligence play in achieving environmental sustainability and what kinds of policy measures should be taken to ensure Environmental Justice is completely accounted for when developing solutions to our climate crisis with AI?
PS: We have seen many applications (monitoring environments remotely, enabling better integration of green or independent sources into grids) which highlight the many potential applications that can help us meet our environmental goals at a faster pace. Remote monitoring and prediction based on it, for example, can directly help inform policy on deforestation, desertification or other macro trends. However, it is also important to acknowledge the environmental burden that compute resources for large models create. Large ML [Machine Learning] models can have a carbon footprint of hundreds of thousands of kg of CO2 so we do need to balance this with their potential benefits, ensuring we are conscious of their impacts.
EH: What’s a fun fact about you that not many people know about you?
PS: I grew up horse riding and competing in dressage and show jumping for a few years in Brunei. That’s probably the most unexpected fact!
EH: What would you say to someone who has never participated in a hackathon before?
PS: Everyone at the hackathon was once someone who had never attended a hackathon before so don’t be nervous to try because of a lack of experience! There’s something for everyone so embrace a weekend where you’ll learn a lot while meeting like-minded individuals!