Image
How UVA’s MSDS Helped a Data Scientist Become an ML Operations Engineer
Timothy Rodriguez
Employment
LMI Consulting, LLC, Machine Learning Operations Engineer (Tysons, VA)
Education
M.S. in Data Science 2025, University of Virginia
B.A. in Economics and Political Science (Foreign Affairs) 2018, University of Virginia
Timothy Rodriguez was already a successful data scientist working for a government contractor. But what he wanted was to use machine learning to solve real world problems. This goal led him to pursue the M.S. in Data Science (MSDS) online program from the University of Virginia, where he got the foundation he needed to take his career to the next level.
Rodriguez's experiences with the UVA School of Data Science demonstrate how the master's program enables students to gain hands-on experience, make impactful connections with corporate partners, and leverage their newly acquired skills to become leaders in the field. We spoke with Rodriguez to learn more about his MSDS experiences and how they helped him accelerate his career.
Q: What were you doing before the MSDS program, and how did your career change after graduating?
Before the MSDS program, I was a data scientist at CGI Federal supporting the Office of the Undersecretary of Defense for Acquisition. I was mostly focused on automating integrated data pipelines and executive level dashboards for reporting purposes.
After graduation I found a new role with LMI supporting Army Special Operations Command (USASOC). I will be working to support USASOC with deploying advanced lightweight computer vision models to edge devices in the field. This change is a significant shift in my career toward more hands-on application of machine learning models directly supporting the mission. This is exactly what I hoped the MSDS program would provide me, an opportunity to get involved with applying machine learning to solve real world problems.
Q: How has the MSDS degree accelerated your career or opened opportunities that wouldn’t have been possible otherwise?
The MSDS program allowed me to gain exposure to and knowledge of computer vision and machine learning pipeline creation, neither of which was available to me in my previous roles.
This experience allowed me to pursue opportunities explicitly working with advanced artificial intelligence/machine learning models. It also provided me with the foundational understanding of how these models work to be successful in those roles.
While I already possessed experience with basic data pipelines and visualization from my previous role, the MSDS program took that experience and built on it by providing me with a foundation in ML and AI required to take my career to the next level.
Q: What does a typical day in your role look like, and what drew you to this position within the data science field?
Essentially, I am responsible for making sure that all aspects of our project are properly integrated and operating efficiently. Our project works with a lot of sensor data that is fed into a computer vision detection model, which then allows a decision model to recommend actions to a human in the loop. My job is to make sure that each step in the process is providing the right data in the right format. I also ensure that, when required, we can retrain our current model or swap it out for a new one quickly and efficiently.
Since our work relies on speed and efficiency to be effective, my job is vital to ensuring the mission is accomplished. I was attracted to this position since it involves an opportunity to work with all aspects of the data science process, creating pipelines, cleaning data, training/retraining models, troubleshooting connection issues, as well as designing useful visualizations for quick decision making.
Q: Can you share a specific project or problem at work where you directly applied skills from the MSDS program?
One of the key things I was responsible for in my previous data science role was creating and managing integrated data pipelines that aggregated data from across the Department of Defense to create actionable insights for leadership. Near the end of my time in that role, we inherited the process of building and maintaining API (application programming interface) connections between the data lake platform Databricks and authoritative source systems. Professor Kropko’s DS6001 course (Data Engineering I) covered both API calls and data pipeline optimizations, which allowed me to take control of the existing pipelines, optimize them as well as build new ones from scratch. The organization we inherited the work from had previously taken months to develop these pipelines. With my knowledge from the MSDS program, I was able to create similar pipelines significantly faster.
Q: How did your MSDS capstone and/or internship experience prepare you for the work you’re doing now?
Interestingly, my capstone sponsor and my current client share some striking similarities with respect to resource and compute restraints. My capstone sponsor, the Renewable Energy Wildlife Institute, presented a problem to our group in which they needed to classify research papers into a myriad of tag groupings based on the content of the paper, and they needed to do so within a very resource-constrained environment. As a nonprofit, they did not have the budget to use enterprise-level classification models or retrieval-augmented generation (RAG) pipeline tools, so we had to get creative with lightweight models that could be run locally on a laptop to achieve that task. This is very similar to the edge device resource limitations that exist within my current role supporting Army Special Operations Command. While many clients have access to large data centers and significant compute resources to train and deploy models, many are constrained to very limited hardware environments where efficiency and speed are key. My experience working within the REWI projects resource constraints exposed me to model quantization (essentially shrinking a model down to be able to run on a less powerful machine), which is a technique I heavily leverage in my current role.
Q: What made UVA’s MSDS program stand out to you? Were there specific classes, projects, or professors that were especially impactful?
I would say that the program design and structure is what stands out the most. The progressive scaffolding of complexity and difficulty throughout the semesters allows students to build their foundations together while still offering challenges to those students further along. The constant evolution of the course material and elective offerings helps keep the program feeling fresh and relevant in the extremely dynamic landscape of data science. How to Train Your LLM delved into increasingly relevant technologies and Healthcare Data Science taught us about ever important industries where data science has tremendous potential to enhance and alter the space.
Finally, the professors we have the privilege to work with are second to none. They are experts in their field, genuinely enjoy teaching, and provide a level of engagement with the material that made this program a truly impactful experience.
Q: What’s exciting to you within the industry right now? What trends, innovations, or breakthroughs in the industry are you paying attention to?
The most exciting things to me right now are the breakthroughs and improvements with respect to both edge computing and model quantization. The ability to shrink once-massive models that take days if not months to train on very powerful computers into small efficient versions of themselves that can run on phones and laptops is going to hugely increase the usability of all kinds of models. This is what I like to call the democratization of AI.
I think this will create the biggest impact for the nonprofit, research, and humanitarian spaces, which are very resource constrained compared to big tech and government. Getting powerful but efficient models down to the people on the ground, the subject matter experts who know how to use them, will enable much more impactful leveraging of AI for the common good.
Q: What UVA experiences (networking, alumni support, projects, mentorship, etc.) were most influential in shaping your career exploration and search process?
The networking opportunities both on Grounds at the new Data Science building as well as in Northern Virginia were the most impactful from my perspective. Those opportunities allowed me to connect with my fellow MSDS classmates and potential employers within my industry and beyond. My current position was obtained through a referral via my extended network. My employer was well aware of the MSDS program through other individuals they knew who had gone through the program.
Q: How do you stay connected to UVA and the School of Data Science, and why does it matter to you?
I stay connected to UVA and the School of Data Science through attending events like Datapalooza; engaging with professors, classmates, and current students; and keeping up with research via social media outlets like LinkedIn. Connecting with the UVA community both expands and maintains my personal and professional network while allowing me to give back to the community that provided so much to me during my time on Grounds.
Q: What advice would you give to prospective MSDS students?
Ask questions and stay curious. This advice goes for prospective students and current students alike. The admissions team was extremely helpful when I was researching programs and deciding where to apply and ultimately attend. If you are unsure about something, ask them. If you do end up in the program, never stop asking questions. There is a decent chance you will never have as much access to some of the smartest minds in the country as you will in this program, so take full advantage of that opportunity. I can also guarantee that if you are confused about something in class, someone else is too.
Learn more about the part-time, 100% online M.S. in Data Science at the University of Virginia. Request more information, connect with Admissions, or start your application today.
