Amstat News regularly profiles government statisticians. For this issue, we feature two leaders from the statistics group at Los Alamos National Laboratory (LANL). Kary Myers is the current group co-lead. Joanne Wendelberger started at LANL in 1992 and has been project leader, deputy group leader, group leader, and acting deputy division leader. She recently returned to full-time technical work as a senior-level scientist.

Joanne R. Wendelberger has been a statistician at Los Alamos National Laboratory since 1992. She holds a bachelor’s degree in mathematics and economics from Oberlin College and master’s and PhD degrees in statistics from the University of Wisconsin-Madison. Her research interests include design and analysis of experiments, statistical intervals, errors and uncertainty, analysis and visualization of large-scale simulations, and education modeling. She has been active in ASA chapters and sections, conference organization, and editorial service. Wendelberger is a fellow of the American Statistical Association and a recipient of the American Society for Quality William G. Hunter Award.”

Please tell us about your work at LANL, including how you came to work at LANL, the positions held there, and your prior training.

After completing my master’s degree, I worked as a statistical consultant at General Motors Research Laboratories. Exposure to a variety of statistical problems arising in automotive applications motivated me to return to Wisconsin, where I completed my PhD in statistics working with George Box. While working on my dissertation, I saw an ad in Amstat News for a statistician at Los Alamos National Laboratory. I wasn’t really looking for jobs yet, but the position looked like a unique professional opportunity located in the mountains of Northern New Mexico. I decided to take a leap and have remained at LANL for 28 years as a technical staff member, project leader, deputy group leader, group leader, acting deputy division leader, and senior-level scientist.

What do you like most about working as a statistician at LANL? What’s most challenging? What work/accomplishments are you most proud of?

The LANL Statistical Sciences Group develops and applies statistical methods motivated by interdisciplinary challenges. I enjoy working in collaboration with amazing colleagues in statistics and other fields. Having a group of about 30 statisticians means there is always someone down the hall to help bounce ideas around. Some of my most impactful work has resulted from analysis and visualization of large-scale simulation models and design and analysis of aging experiments for materials and components. My greatest personal satisfaction has come from serving as group leader of the LANL Statistical Sciences Group, where I could support my fellow group members in their technical endeavors and careers. One of the projects I am most proud of is a current collaboration with researchers at The University of Texas at Austin on an educational research project focused on understanding factors that influence success in pursuing STEM degrees and careers.

What advice would you offer younger statisticians and students who might be interested in working for the government?

There are abundant opportunities for statisticians to contribute statistical skills to different types of statistical problems and applications in the government sector. Working in government provides opportunities to contribute to issues that affect people in many ways. Look beyond the immediate tasks and see how you can connect to broader efforts and longer-term opportunities. I have always advised students and early-career statisticians that a job is what you make it and to think about your job not just as a job but as part of a lifelong career. I have found that being involved in the ASA and other professional societies has benefited my career immensely. In particular, I encourage statisticians to get involved in sections and interest groups that can provide a technical/professional focus, networking opportunities, ongoing learning, and other benefits.

How has your statistical work changed over your tenure with LANL, and where do you see statistical work heading in an era of machine learning, AI, and data science?

After progressing from purely technical roles to a series of R&D leadership roles, I transitioned back to full-time technical work with the perspectives of both a researcher and a manager. I have witnessed a change over time from small focused projects to larger, more complex challenges. I see a trend toward more integrated approaches to problem solving with increasing interaction across both the scientific components and the different aspects of data collection and analysis. I currently serve as the analysis lead for an analytics project at LANL where operational data streams are integrated into a variety of analyses to support data-informed decision-making. I see statistics continuing to provide a rigorous foundation for many of the methods used in machine learning, AI, and data science, as well as a structured approach to uncertainty, while other disciplines bring additional concepts and tools for analysis, computation, and efficient work flows.

Please tell us about your personal interests and hobbies.

I enjoy daily opportunities for exercise—including walking, hiking, and swimming—that help me immensely both physically and mentally. I find that cooking provides endless opportunities for creativity and experimentation that draw upon my statistical background and knowledge. I have traveled extensively with destinations providing cultural experiences, adventure, and the opportunity to help others, including Mount Kilimanjaro, the Himalayas, Mount Fuji, the Galapagos Islands, and Cameroon. Most of all, I enjoy spending time with my family, including my husband and fellow statistician, Jim, whom I met in graduate school; my three daughters, Barbara, Beth, and Laura—two of whom are statisticians; and my granddaughter, Genevieve. JSM has been a professional and family destination for decades!

Kary Myers is a scientist and deputy group leader in the Statistical Sciences Group at Los Alamos National Laboratory. At Los Alamos, she has been involved with a range of data-intensive projects, from examining electromagnetic measurements to aiding large-scale computer simulations to developing analyses for chemical spectra from the Mars Science Laboratory Curiosity Rover. She is an ASA Fellow and has served as an associate editor for the Annals of Applied Statistics and Journal of Quantitative Analysis in Sports. She created and organizes CoDA, the Conference on Data Analysis, to showcase data-driven research across the Department of Energy.”

Please tell us about the Statistical Sciences Group at Los Alamos National Laboratory and the projects you work on.

Our group has almost 30 scientists—I love saying “I’m a scientist!” —most with a PhD in statistics. We’re very much an applied group, by which I mean our work typically starts with a consequential scientific challenge, such as predicting the impact of hurricanes on the power grid, estimating the chemical composition of rocks on Mars, forecasting COVID-19 cases and deaths, or evaluating the state of the nuclear stockpile. We then explore the data sets—ranging from massive to meager—that we can use to address that challenge, and we determine what methods will be appropriate. We’re extremely data driven, whether those data are experimental or simulated or just what our collaborators have lying around. Sometimes, standard methods will do the trick—it’s surprising how far linear regression can get you—but our data violate model assumptions a lot, so we make modifications or develop new approaches.

Tell us about your training and career path before arriving at Los Alamos.

I started my career at Carnegie Mellon, where I earned my undergraduate degree in statistics with a minor in computer science. That’s when I began doing collaborative, interdisciplinary research, including working with a team of statisticians, computer scientists, and astrophysicists to study the Sloan Digital Sky Survey. The faculty recruited me to join a new joint PhD program between the statistics department and what later became the world’s first machine learning department. After winning a fellowship from AT&T Labs to support six years of graduate work, I did summer internships in AT&T’s artificial intelligence department, their machine learning department, and then at a machine learning startup. For my PhD thesis, I worked with large data sets to combine video of the surface of a brain with physiological measurements like blood pressure and heart rate. Knowing I didn’t want a career in academia, I applied to become a scientist at Los Alamos instead.

What do you like most about working as a statistician at Los Alamos? What’s most challenging?

In the best jobs, you work on projects you’re excited about with people you enjoy being around, and I get both here. We tackle amazing problems with some of the top scientists and best computing resources in the world. The people in our group are always keen to hear about your work and help you brainstorm ways of approaching your latest challenge. I often say that when someone hires one of us to work on their project, they’re actually gaining the expertise of our entire group.

The group has several definitions of success. We value publishing and proposal writing but don’t require either. Some people are entrepreneurial about creating opportunities to pursue the work they want to do. Others are tremendous at lending their statistical knowledge to support some of the most critical missions in the nation. The challenge lies in ensuring that such a diverse group can thrive.

Tell us about your position as leader of the Statistical Sciences Group at Los Alamos National Laboratory. What about this position appealed to you, and what are your priorities for the next few years?

I’m our deputy group leader, which means I help with everything from hiring decisions to student programs to sharing our group’s accomplishments with a larger audience. I’m also still a scientist and have several technical leadership roles, including guiding teams studying Mars or nuclear reactors.

I’ve enjoyed this role because I get to support and showcase so many smart people working on such a wide range of projects and initiatives. It also has enabled me to continue growing CoDA, the Conference on Data Analysis, a biennial event that Earl Lawrence and I founded in 2012 to feature data-driven research across the Department of Energy.

Over the next few years, I’d like to expand and formalize our academic connections—including having faculty visit for a week or a summer or a year—and continuing to engage with students through internships or by providing challenging data for their thesis work.

What makes someone a good candidate to work in your group or at another national lab as a statistician?

We love to see people who have worked collaboratively on applied problems and made the effort to know the subject matter as much as the statistical methods. If you’ve worked with a variety of challenging data sets and can articulate what made them challenging and how that steered your methodology decisions, you’ll fit in well. Computing skills will help you stand out as a strong candidate, whether that’s developing an R package or deploying your analyses on a supercomputer. And being able to communicate your work—both the application area and your methodology—to a smart audience that doesn’t have your expertise will make you a coveted collaborator.

To join our group at Los Alamos, you’ll need to be able to obtain a Department of Energy security clearance that usually requires US citizenship. Some national labs have more flexible citizenship requirements than we do, so check each job posting for details.

How does one find out about opportunities for internships and employment at LANL?

We are hiring, pandemic or no pandemic! And this summer, we’re learning how to do our interviews online, instead of onsite. To see the job posting for our scientist positions and learn more about our group, visit the LANL website.

If you’re thinking about an internship, contact people in our group who are working on topics that interest you. The staff profiles at stat.lanl.gov are a good place to start, as is browsing our Google Scholar pages to see what we’re publishing. Many of our current group members started out as summer interns, so I encourage you to talk with them to learn more about their experiences and to hear about beautiful northern New Mexico!