pymc-devs · reshamas · Jun 6, 2025 · Jun 6, 2025 · Jun 6, 2025 · Jun 6, 2025
diff --git a/_static/gsoc_2025/michal.png b/_static/gsoc_2025/michal.png
diff --git a/blog/blog_gsoc_2023.md b/blog/blog_gsoc_2023.md
@@ -49,7 +49,7 @@ Expand support for spatial models in PyMC
 
 Find out more at [Daniel's GSoC blog](https://daniel-saunders-phil.github.io/imagination_machine/).
 
-1. __What motivated you to apply for the intership with PyMC?__ 
+1. __What motivated you to apply for the internship with PyMC?__ 
 
      A few years ago, I started reading Richard McElreath’s marvelous book Statistical Rethinking and fell in love with probabilistic programming. My prior coding experience was in Python so PyMC was a natural choice of probabilistic programming language to pick up. Since then, I’ve really grown to love the package and wanted to learn how to contribute to it in a serious way.
 

diff --git a/blog/blog_gsoc_2025.md b/blog/blog_gsoc_2025.md
@@ -0,0 +1,83 @@
+(interns_2025_presentations)=
+# Meet our 2025 GSoC PyMC Interns
+
+:::{post} June 6, 2025
+:tags: gsoc, community, contributing
+:category: news, testimonial
+:author: Reshama Shaikh, 2025 PyMC Interns
+:image: 0
+:exclude:
+
+We are excited to introduce to the community our cohort of 2025 Interns working on PyMC.
+:::
+
+
+## Michal Novomestsky
+
+**Project Name**
+
+Implementing Integrated Nested Laplace Approximations (INLA)
+
+**Project Description**
+
+>A key component of Bayesian inference is integrating over prior distributions to obtain posteriors. In practice however, these distributions are often high-dimensional, resulting in a significant computational cost associated with integration, which remains a key challenge in Bayesian ML. Under certain assumptions, it is possible to efficiently compute posteriors for Latent Gaussian Models (LGMs), which represent a broad class of statistical models in Bayesian statistics. This is known as the method of Integrated Nested Laplace Approximations (INLA), and this project aims to implement a working basis for INLA in the PyMC library.
+
+**Mentors**
+- [Rob Zinkov](https://github.com/zaxtax)
+- [Colt Allen](https://github.com/ColtAllen)
+- [Theo Rashid](https://github.com/theorashid)
+::::{grid}
+:::{grid-item}
+:columns: 12 6 6 4
+
+![photo of Michal Novomestsky](../_static/gsoc_2025/michal.png)
+:::
+:::{grid-item}
+:columns: 12 6 6 8
+
+**Bio**
+>I am a 4th year undergraduate student at Monash University in Australia, studying Aerospace Engineering (Honours). Although I come from a physics/engineering background in fluids, I have since shifted toward ML/AI as I have found some of the research directions in this space very exciting. My research interests can be summed up as “how can we embed priors/known information into AI/ML models?” and “how can we get them to reason and provide guarantees about their output?”. Bayesian ML is a promising direction to address these questions and is a key interest of mine.
+
+**Connecting**
+- Website: [michal-novomestsky.github.io/](https://michal-novomestsky.github.io/)
+- LinkedIn: [@michal-novomestsky](https://linkedin.com/in/michal-novomestsky)
+- GitHub: [@Michal-Novomestsky](https://github.com/Michal-Novomestsky)
+:::
+::::
+
+
+1. __What motivated you to apply for the internship with PyMC?__ 
+
+     As mentioned in my bio, I am interested in pursuing research focused on developing more principled and structured AI models, as well as understanding the internal behaviour of existing architectures. Bayesian statistics provides potential inroads to address both these challenges. As such, I felt that this would be an excellent opportunity to better understand the field as well as network with experts.
+
+1. __Why did you choose your specific project topic?__
+
+     I have a taste for more fundamental research in AI, largely focused on the math underpinning it. Of the projects on offer with PyMC, INLA seemed to fit this description best - a fundamental technique used to estimate posteriors - in other words, it would be a project with a theoretical focus much closer to the models themselves, rather than focusing on more backend aspects.
+
+1. __How did you get involved in open source software?__
+
+     Having written an academic paper in the past, I have previously developed tools for data analysis and denoising which are publicly available on my GitHub.
+
+1. __What are you expecting or hoping to get out of your internship experience?__
+
+     I am aiming to develop a better understanding of Bayesian statistics not only from a theoretical perspective, but also to better appreciate some of the technical challenges it faces in practice. I also hope to gain invaluable connections from this project.
+
+1.  __What are your career goals? How do you see the internship program moving you towards them?__
+
+     I am hoping to pursue a career in research, in particular focusing on developing more robust and interpretable AI, and GSoC is an excellent opportunity to be able to network with experts across both industry and academia, which I hope will open doors for me to continue to pursue this passion.
+
+
+
+
+## References
+
+- [2025 GSoC NumFOCUS projects](https://summerofcode.withgoogle.com/programs/2025/organizations/numfocus)
+- [2025 PyMC GSoC applications blog](https://www.pymc.io/blog/blog_gsoc_2025_announcement.html)
+- [2024 GSoC NumFOCUS projects](https://summerofcode.withgoogle.com/programs/2024/organizations/numfocus)
+- [2019 NumFOCUS GSoC announcement](https://numfocus.org/blog/meet-our-2019-gsoc-students-part-2)
+
+### GSoC Interns Announcements
+- [2022 GSoC PyMC blog](https://www.pymc.io/blog/blog_gsoc_2022.html)
+- [2023 GSoC PyMC blog](https://www.pymc.io/blog/blog_gsoc_2023.html)
+- 2024 GSoC PyMC blog: none
+