Purdue University

• Overview
• How to Cite Anvil
• Training
• Documentation
• Anvil Support Hour
• Research Experience for Undergraduates
• Scientific Highlights
• Contact Us
• Anvil in Action
• Why HPC?
• Advisory Board

Meet Your Mentor!

We sat down with our 2024 Anvil REU Mentors to discuss their role at RCAC, what their REU students will be working on, and much, much more. Keep reading below to learn more about your mentor.

 

Jungha Woo

Please introduce yourself:

My name is Jungha Woo, and I am a lead software engineer working at Scientific Software Group at the Rosen Center for Advanced Computing.

 

What do you do?

I help researchers run their code faster and easier on some Purdue or National clusters. Sometimes, I develop software running on the science gateways for researchers to share their experiment/simulation tools with colleagues. For researchers who want to integrate their models with others, I containerize models and automate the whole workflow for easier model coupling. When researchers want to apply AI/ML techniques, I work with them closely to analyze the problems, provide solutions, implement machine learning models, and run the prediction models on Purdue clusters.

 

Why would I come to you for help?

If you are interested in developing research software and curious about optimizing the code on the clusters, I will be delighted to answer your questions. I have worked on high-performance computing since 2006 and have had experiences with diverse domains, including Computer Science, Geospatial processing, Hydrology, Agricultural economics, Mathematical Finance, Statistics, and Artificial Intelligence/Machine learning.

 

What’s one professional skill you’re currently working on?

I'm working on Reinforcement Learning and Natural Language Processing with Deep Learning.

 

Why did you decide to become a mentor for the REU program?

I wanted to develop some Open OnDemand custom apps, as this will significantly improve users' experience submitting and managing jobs. Developing apps for the Gromacs Gateway would be a good topic for undergraduate students to get familiar with HPC software, learn software development skills, and contribute to the research community. I have had great opportunities to advise a few undergraduate and graduate students on some Jupyter Notebook development. I believe those experiences will help undergraduate students achieve their goals during the REU program. 

 

What do you hope the REU student will get from this experience?

I hope the REU learns how to tackle problems independently. Often, REU students get frustrated about where to start. Trial and error is one of humans' oldest methods, but it is one of the most effective ways for beginners. I hope students learn how to describe their problems well, google similar issues, find solutions by reading manuals or articles, and finally resolve the issue after many trials and errors.

 

How can this help them in their college career?

Independent problem-solving skills are one of the most essential skills any worker should possess. Anyone can apply this skill to general problem-solving procedures during college, graduate school, or an industry career.

 

How can this help them in their professional career?

Professionals do research all the time. They google to find solutions when running into issues. Such problem-solving skills help you figure out the issue quickly. If the problem is well-known, you can get an immediate solution. However, if it is unique or rare, you must study again. This skill gradually shortens the time to explore new topics, understand root causes, find eligible solutions, and improve efficiency.

 

Why HPC?

    1). As an industry to work in?

    HPC has become a necessity for graduate students and faculty to do research. You can work with diverse faculty and domains as a research software engineer at HPC. It is fun to understand problems and solve them with what you learned from school: algorithms, design patterns, statistical analysis, and so on. This job is for you if you are thrilled to solve challenging problems and love to learn.

    2). As a use for research?

    You will work on advanced simulation/experiment code developed by one of the top-level research scientists or faculty at Purdue University. You will develop software tools for such cutting-edge research. That software will be published online for the public, benefiting many new users (students and researchers). You can make broad impacts on the research community.

    3). As an experience for an undergrad?

    Typically, undergraduate students have few chances to learn what graduate students and faculty do for research. Since you come across diverse domains using HPC, you will know what problems in the research fields are being investigated, how they collaborate through sharing experiment results, what bottlenecks there are in terms of software development, and what you need to study or prepare if you want to dive into that area.

 

Why RCAC/Purdue?

    1). As a place to work as a professional?

    You will get the maximum degree of freedom in solving problems. You can develop the best data structure/algorithm you think and apply your idea directly to the software frequently used for research. Also, there are many knowledgeable colleagues you can ask for help. Professionals have diverse backgrounds in computer science, electrical engineering, chemistry, physics, biology, etc.

    2). As a place to get your education?

    One of the top ten engineering schools in the U.S.

    3). As a place towork as an undergrad?

    Suppose you want to pursue a grad school degree. In that case, there is a probability that you may get familiar with the software developed by the faculty you are interested in, as most faculty members use Purdue HPC/Research Computing. Take advantage of every chance you can use to promote yourself, including software development experience in the HPC area. 

    

Where do you see your specialty in HPC going in the next 10 years?

I want to help more researchers who want to apply AI/ML technic to their domains. Nowadays, it is not uncommon for a graduate student in chemistry to develop a better classification algorithm using machine learning and contribute to the AI/ML community. Also, some graduate students in Civil engineering are working on developing energy management systems for smart buildings using Reinforcement learning. HPC/SuperComputing/Research Computing has solid needs for AI/ML experts who can consult faculty on grant writing and AI/ML model development.

 

Have you ever been an REU or student employee?

    1). What was your experience?

    I did a graduate internship at IBM Research.

    2). What did you learn?

    How to study unfamiliar topics, working together as a teammate.

    3). How did this propel you into your current job?

    Understanding other researchers' problems and approaches was very interesting, and it was one of the necessities of my current job.

    4). What will you do differently as a mentor for students to make the experience better?

    I would encourage them to ask as many questions as possible. Understanding the problems is the first step toward finding solutions. Don't be shy to ask questions. You need to know what you don't know to make progress.