GSoC 2025 | Beginner Interested in State Space Models Project

Hi everyone,

I’m Krishna Vallabha Goswami, an undergraduate student currently exploring the world of Bayesian inference and time series modeling.

I recently discovered PyMC’s GSoC 2025 project ideas, and I’m particularly interested in the State Space Models project — especially the part about implementing dynamic factor models. I’m still a beginner in contributing to open-source and in deep time series modeling, but I’m extremely excited to learn, contribute, and grow through this experience.

Over the past couple of days, I’ve been reading through the PyMC documentation, exploring the pymc_experimental.statespace module, and trying to understand how SARIMAX and similar models are built internally. I also plan to go through PyTensor in more detail to understand the backend better.

I’m currently drafting a proposal for GSoC and would really appreciate any guidance, feedback, or suggestions. I’m eager to put in the time and effort to make a valuable contribution.

Thanks so much for this opportunity!

– Krishna

Thanks for your interest in GSoC, Krishna. Time is getting rather short, as the deadline is tomorrow at 18:00 UTC. But, if you do have a proposal that is ready to go feel free to pass it along. Have you worked with our code base before (i.e. submitted a PR)?

Thank you for your response, Chris! Yes, I’m aware of the deadline and I’ve already submitted my full proposal via the GSoC portal. I haven’t contributed a PR yet, but I’ve been actively reviewing the codebase and plan to submit one shortly to continue engaging with the community ,regardless of the outcome. I’m genuinely passionate about this project and excited about contributing more to PyMC.