Internship Description
Sebastian Steiner, '23SEAS, interned with Tamer Fund for Social Ventures portfolio member Plentify, a sustainable energy startup in Cape Town that aims to enable South Africans to switch to reliable and sustainable energy sources while minimizing utility bills by harnessing the power of AI and advanced grid technology. As a data science intern, Sebastian worked on data-driven projects to predict the impact of water heater load management on energy costs, detect anomalies in water heater performance, and make design decisions on prototypes of new products.
During my summer at Plentify, I worked on two primary projects. The first project focused on developing a solution that mitigated power spikes after load shedding events, which are scheduled power outages that occur multiple times each day because the energy supply is insufficient to cater to the entire population of South Africa simultaneously. After load shedding, many household appliances will turn on at once causing a power spike, which damages the electrical infrastructure and potentially induces a second outage. My project researched the optimal strategy to stagger the activation of appliances without compromising customer experience.
I also analyzed the customer savings when shifting the energy required for heating water tanks. Water heaters typically consume energy during peak hours, which are in the morning pre-work and the evening post-work. In these periods, the energy sources are less eco-friendly. My analysis investigated the benefits of transitioning the energy load from peak hours to midday, when surplus solar energy is available. This strategic shift not only contributed to a cleaner energy mix but also opened avenues for the development of innovative features that optimize solar energy utilization.
Plentify is focused on making energy supply affordable, reliable and clean using smart technology. During my internship, I helped Plentify offer a more reliable energy supply through my project in managing power after load shedding, and develop mechanisms to source cleaner energy, via my solar savings analysis.
As a masters student in data science at the Fu Foundation School of Engineering and Applied Science, I found it hugely beneficial to intern as a data scientist because I was able to both apply and develop the skills I had learned in the classroom. Particular classes that I found helped me in my internship included Exploratory Data Analysis and Visualization. In this class, I studied the art of communicating through data, which was extremely useful when I shared my project work with my supervisors and managers. Another class that I found beneficial was Applied Machine Learning. This was a hands-on course where I practiced working on data projects end-to-end. That is, how to clean raw data, preprocess it, use it to build models, evaluate the performance of the model and vitally how to document these steps clearly. I also hope to apply the skills that I developed during my internship to support my studies at Columbia.
One of the biggest challenges I encountered during my internship was understanding the extent and disruptiveness that load shedding causes. Load shedding are scheduled power outages that occur daily because a nation does not have enough electrical energy to supply its population all at once. As I have always lived in countries without load shedding, I did not fully appreciate the detrimental effects that it has on a country's economy and an individuals'daily life routines. Through my visit to Cape Town and project work, I had firsthand experience with load shedding events, that in extreme cases last for six to eight hours a day for weeks on end. I learned about its impact on the population on a granular level — from restaurants not being able to carry out business activities and serve customers, to having to wait several hours to have a hot shower in the evening, and to having no lights or heating during the day in the middle of winter. From these experiences and project work on developing products to mitigate the impact of load shedding, I now truly appreciate the importance of finding solutions to this issue.
I was incredibly fortunate to have the opportunity to intern at Plentify through the Tamer Center for Social Enterprise Summer Fellowship. I received excellent mentorship from my supervisors, which will certainly help me on my journey to becoming a data scientist. Being able to take the data science skills that I learned in the classroom at Columbia University and applying them in real-world situations was extremely rewarding because I could see the impact these data science skills have on businesses. Moreover, I look forward to translating the skills that I developed during this internship back to my studies at Columbia University, specifically my Capstone project, and forward into my professional career.
Another highlight of my internship was being welcomed into a startup environment. The exciting and accelerating growth of Plentify, within South Africa and internationally, fuelled a dynamic and vibrant workplace. I enjoyed working in this ambitious environment that was making an impact on the energy crisis in South Africa.
One of my biggest takeaways was the importance of Plentify's mission to make energy cleaner and more reliable in developing countries and countries with rapidly growing populations. In these places, the infrastructure is not sufficient to supply the entire population with energy and it is typically too expensive to carry out the necessary upgrades to improve the electricity system. This leads to erratic energy supply and the use of carbon-intensive diesel generators. However, through the use of smart technology that controls consumer energy demand, pressure on the electrical can be reduced, which enables a more consistent energy supply, and the use of renewable energy sources can be leveraged to reduce the carbon footprint.
