Northwestern truly embodies the essence of collaborative work.”
Akash Pandey
PhD Candidate in the Department of Mechanical Engineering
Akash Pandey is a PhD candidate in the Department of Mechanical Engineering in the McCormick School of Engineering. He develops explainable deep-learning models to study and predict protein behavior, uncovering the drivers of traits like stability and strength. Akash’s research spans machine learning, solid mechanics, and biomaterials design.
How would you describe your research and/or work to a non-academic audience?
Deep-learning (DL) models are usually thought of as black boxes, which leads to a lack of trust in their predictions. In my research, I develop explainable deep-learning models to study and predict protein behavior. By incorporating explainability, I uncover the underlying reasons for specific protein properties, such as high stability or enhanced mechanical strength.
What have been some of the most memorable twists and turns of your career?
Although I am currently working in machine learning, my academic journey began with an undergraduate degree in automobile engineering, followed by a master’s in solid mechanics. After my master’s, I worked at Rolls-Royce, an aerospace company. In the third year of my PhD, I interned at Capital One as a data scientist. In short, my career seems to have taken a new turn every 2-3 years, and honestly, I’m glad it did—each step has shaped my journey and led me to where I am now.
Tell us what inspired your research and/or work.
While working at Rolls-Royce, I began developing an interest in data-driven models. To catch up with the rapidly evolving world of machine learning, I took several courses on Coursera and became especially interested in sequence models. Then, at Northwestern, when I got an opportunity to work with Professor Sinan Keten and Professor Wei Chen on proteins, I jumped at this opportunity. What drew me in was the fact that the protein's primary structure can be defined as the sequence of amino acids, making it a perfect fit for applying machine learning-based sequence models.
Whom do you admire in your field and otherwise, and why?
I deeply admire the work of both of my advisers, Professor Sinan Keten and Professor Wei Chen. Professor Keten is well known for his contributions to the computational modeling of proteins and polymers. I admire not only his strong command of the field but also his curiosity and openness to exploring new areas. During our discussions, he consistently asks thoughtful questions that significantly enhance the quality of the work. Professor Chen is highly respected in the field of design, and her quest for developing innovative new methods to solve the problem is inspiring. She encourages her students to approach research with a "design mindset," ensuring that the methods we develop are not only useful for predicting properties of existing materials but also for enabling the design of entirely new ones.
Last but not least, I am very inspired by my wife, Payal Mohapatra, who is also a PhD candidate in the Electrical and Computer Engineering Department at Northwestern. Though she is not from my field, she has been my greatest source of motivation. She consistently inspires me to aim higher. One of the things she often says is, “If you’re feeling comfortable where you are, it means you’re not aiming high enough.”
What is the biggest potential impact or implication of your work?
The explainable deep-learning model, named COLOR, that I developed for proteins can identify critical motifs (segments) in proteins that are mainly responsible for modulating their properties. We are currently working to demonstrate how these motifs can accelerate the design of environmentally friendly, protein-based biomaterials.
Why Northwestern?
Northwestern truly embodies the essence of collaborative work. If you work on machine learning methods, you can easily find collaborators who do experiments or simulations and are willing to collaborate. I strongly believe that collaborative work brings out much deeper insights.
How do you unwind after a long day?
I like going for long walks with my wife. If the temperature is favorable, I also go for a quick run.
What inspires you?
Working around colleagues who are also trying to solve interesting problems in their work.
How would your closest friends describe you?
A calm person!
What did you originally want to be when you grew up?
I wanted to be a cricketer and played the game tirelessly. Even before important exams, I would find time to play, as it always filled me with positivity.
What advice would you give your younger self or someone considering a similar path?
Never be too proud to take advice. Keep your mind open, be willing to absorb feedback, and act on it. Do not wait for the motivation to kick in; put in the work every day. Over time, you’ll look back and realize just how far you’ve come.
Tell us about a current achievement or something you're working on that excites you.
My current work on the explainable machine learning model (named COLOR) for proteins was accepted at the ACS Journal of Chemical Information and Modeling. I am proud of this work as this is my first methods paper, and it was well received by the community. We are now working to extend its capabilities to accelerate protein design through targeted mutations. Additionally, we are expanding COLOR’s scope beyond proteins to other time series applications, such as human activity recognition.
Tell us about a time when things did not go as you planned. What did you learn?
After my undergraduate studies, I had hoped to go and work at a top automobile company, but I was unable to secure a job in my dream company. Rather than settling for just any job, I chose to continue learning and deepen my knowledge in solid mechanics. That decision led me to pursue a master’s degree, and my passion for learning has only grown since then.
What are you most proud of in your career to date?
There is no one thing in particular. However, getting an opportunity to pursue a PhD at Northwestern with such great minds is surely at the top of my list.
Publish Date: July 22, 2025
If you know a graduate student, postdoctoral scholar, graduate faculty member, staff member, or a member of our TGS alumni population who would make a great candidate for our TGS Spotlight Series, please complete this brief TGS Spotlight Series Nomination Form.