**Fig 1.** Introduction to Neural Networks talk at Universidad del Norte, Colombia. May 17th, 2025.
During my undergraduate studies in Industrial Engineering, I took some subjects I didn’t enjoy at all and others I felt more affinity with, but the one I connected with immediately was statistics. The way it allows us to understand the world through probabilities and its applications in different scientific and life contexts deeply caught my attention. In addition to this, I have always felt drawn to teaching, as I believe it’s one of the most beautiful and efficient ways to contribute to society—by sharing knowledge and influencing the professionals of tomorrow. With my personal aspirations and professional goals in mind, I decided to enroll in the Master's in Applied Statistics at Universidad del Norte.
Over these two years, I have undergone both a personal and professional transformation. I entered the program eager to deepen my knowledge of statistical tools to solve real-world problems, and I leave with a more critical, rigorous, and passionate view of data analysis, along with the future goal of becoming a research professor. My research line focused on Topic Modelling, particularly the **Latent Dirichlet Allocation (LDA)** model. This choice was born from my interest in discovering hidden patterns and topics in large volumes of text—something increasingly relevant in the age of unstructured data. As part of my thesis, I developed the guide
***"A Comprehensive Guide to Latent Dirichlet Allocation: Building a Solid Foundation with Key Statistical Concepts"*** (See here), where complex theoretical concepts such as Bayesian inference, Gibbs sampling, and Dirichlet distributions are broken down and linked to the practical implementation of the model. This work, besides being an academic contribution, allowed me to strengthen skills in technical writing, R and Python programming, and results visualization. I also had the opportunity to apply these skills in real-world settings, such as text analysis involving the Marshall-Olkin Bivariate distribution and consulting projects in the port sector, where I integrated machine learning and multivariate analysis.
My time in the master’s program wasn’t just technical. I also served as the student representative on the curriculum committee, which gave me an insider view of the collective construction of the program and allowed me to develop academic leadership skills.
Today, I look back with gratitude for every challenge, every study night, every class, and for the support of professors and classmates who always strived for excellence. This experience has profoundly shaped my career, opened new doors, and consolidated me as a professional with a solid foundation in statistics, equipped with tools that will surely help me achieve my goals.
# **Supervisors' Reflections**
**Fig 2.**
Dr. rer. nat. Humberto Llinás Solano, Coordinator of the Master’s in Applied Statistics and Full Professor in the Department of Mathematics and Statistics, Universidad del Norte, Colombia
Talking about Carlos means talking about talent, commitment, and… lately, someone who came back from Argentina saying things like “che, what a great Bayesian model” with surprising naturalness. From the very first semester, it was clear we had an outstanding student. Academically impeccable and always willing to collaborate, Carlos not only mastered the most complex concepts of applied statistics but was also able to translate them with clarity, style, and even humor.
His work on LDA is a great example of how deep science can be communicated without falling into unnecessary jargon. But that’s not all. Carlos programs in R and Python (and others) like someone playing the piano—with rhythm, precision, and elegance. He created a promotional video for the program that mixed creativity, acting, storytelling, and a dose of charisma we still celebrate. I don’t know if he was more of an actor, journalist, or statistician that day… but I do know he left us all impressed.
He also delivered a talk on neural networks that was simply brilliant: clear, well-structured, and educational. He has a gift for teaching, for connecting with the audience, and for doing it all with a smile and professionalism. As his professor and thesis advisor, I couldn’t be prouder of the path Carlos has taken. He wasn’t just an excellent student—one of the best we’ve had in the program—but also a person of integrity, generosity, and deep humanity.
I’m convinced that wherever he decides to go next, he will continue to excel and leave a mark. Best of luck, Carlos."