Author Spotlight
To Make an Impact in Any Industry, Domain Knowledge Is Critical
Abiodun Olaoye discusses his work on renewable energy
In the Author Spotlight series, TDS Editors chat with members of our community about their career path in data science, their writing, and their sources of inspiration. Today, we’re thrilled to share our conversation with Abiodun Olaoye.
Abiodun Olaoye is an experienced data scientist practicing as a Senior Performance Engineer (data analytics) at RWE Renewables in Austin, TX. He earned a doctorate degree in Mechanical Engineering and Computation from the Massachusetts Institute of Technology (MIT) in 2019. He is a recipient of several academic scholarships, including the Nigerian Presidential Special Scholarships for Innovation and Development (PRESSID), and is a three-time awardee of the University of Lagos Endowment Scholarship.
What was your data science learning journey like?
My interest in data science first surfaced as a graduate student at MIT, during the preprocessing of sensor data from experiments I conducted on ships at the Towing Tank laboratory around 2014. Although at that time I was not formally introduced to data science as it is today, the analyses I did then were essentially time-series analyses.
Perhaps because the focus was more on scientifically investigating the performance of offshore structures and ships in ocaean waves, I did not realize how much of the fundamentals of data science I have built capacity in already. This really helped smoothen my journey in formalizing my data science skills when I started taking courses in Udemy and Cousera. For example, I wrote several programs for data processing and analysis in MATLAB for about 5 years, which made it relatively easy for me to learn Python.
In addition, all my computational science and engineering graduate courses in linear algebra and numerical simulation provided a solid starting point to grow my skills as a data scientist. Hence, my learning curve hasn’t been too steep, to be honest.
Has your approach to data science and machine learning evolved over time as your career progressed?
Definitely. When I started out as a beginner in data science, I enjoyed playing with hyperparameters to improve ML model performance. However, as my experience increased, I spent more time in a given project expanding my domain knowledge to sharpen my feature-engineering skills, while model tuning became the icing on the cake. Hence, my approach gradually shifted from model-centric to data-centric modeling.
In terms of problem-solving, I evolved from solving generic problems with toy datasets to tackling more practical problems with a focus on applying state-of-the-art data science techniques to real-world problems, especially in the renewable energy industry. Looking forward to the future, I see myself extending my application of data science to solving engineering research problems and/or enabling physics-based solutions with data-driven methods.
How has your background in science, engineering, and data played into your current work with renewables?
With a doctoral degree covering the fields of mechanical engineering and computational science & engineering, I have built a broad foundation on which to build my career.
On the one hand, the knowledge of fluid-flow physics, momentum, and energy principles I learned for the mechanical components of my graduate studies formed the basis of my domain knowledge in the renewable energy industry.
On the other hand, the computational aspect of my graduate studies covered advanced linear algebra and numerical solutions of non-linear and partial differential equations, which are relevant to optimization and modeling in data science. Armed with this knowledge, I am able to apply advanced data analytics techniques to generate actionable insights from sensor data to boost the performance of our renewable energy assets.
Do you have any advice to share with others who might want to find sustainability-related work, but aren’t sure how to go about it?
Sure—my candid advice for working in the sustainability field is to first identify which aspects of the industry interest you the most, current challenges, and state-of-the-art solutions being deployed in those areas.
This will help increase your domain knowledge, which is critical to making an impact in any industry. In addition, I recommend building a network of current professionals (through Medium, LinkedIn, etc.) to increase your knowledge of the industry and the chances of landing your first role via shared articles and posts.
You’ve been writing about the intersection of data and renewable energy for a couple of years now. What prompted you to share your knowledge with a broader audience?
There are research and operational opportunities for data practitioners in the renewables industry—our assets generate several gigabytes of data per second worldwide, and we need to better understand this data to improve the performance, availability, and reliability of these assets. I believe my articles can help someone discover a new career prospect or research interest, given the huge available opportunities and the relatively few professionals who possess both data skills and core engineering knowledge.
In addition, I started writing more because I realized that I am able to break down difficult concepts in a beginner-friendly way, which can help ignite the interests of junior professionals while expanding the knowledge of experts.
On a personal note, I started conducting independent research, developing open-source python packages, and writing articles since I first identified the opportunity to help accelerate the world’s energy transition through the application of state-of-the-art data science techniques.
As someone who works in a field that has drawn a lot of attention recently, how do you hope it evolves in the next year or two?
In the coming years, the renewables industry will increasingly rely on modern data technologies to store, extract, integrate, transform, and model sensor data with the overall goal of boosting revenue while reducing the levelized cost of energy through increased turbine performance and availability.
In addition, low-code cloud platforms for technologies, such as AutoML, will help lower the entry barrier for new professionals in the industry. Finally, I would like to see more open-source projects, cross-organization initiatives, and community-driven knowledge sharing which industries (such as information technology) have immensely benefitted from for many years.
To learn more about Abiodun’s work and stay up-to-date with his latest articles, follow his here on Medium and on Twitter. For a taste of Abiodun’s TDS articles, here are a few standouts from our archive:
- 3 Fundamental Processes in Feature Engineering (October 2022, 4 minutes)
- Predicting Solar Power Output Using Machine Learning Techniques (January 2022, 7 minutes)
- Forecasting Wind Power from Multiple Numerical Weather Predictions (July 2020, 6 minutes)
- Clustering-Based Data Preprocessing for Operational Wind Turbines (September 2022, 10 minutes)
- 5 Essential Qualities of Anomaly Detection Systems (October 2022, 4 minutes)
Feeling inspired to share some of your own writing with a wide audience? We’d love to hear from you.
This Q&A was lightly edited for length and clarity.
