EuroPython 2025

Marine Guyot

Embedded system engineer in the automotive industry by training, I have always aimed to work in research. After a detour in the US and attempting to work in the industry, I refocused on my primary objective and moved to Belgium to pursue a career as a research software engineer in bioscience.

I spent three years working for a neuroscience lab that is part of NERF. During this time, I gained a basic understanding of the brain but, more importantly, valuable insights into lab development processes, the goals of research software, and the challenges involved in achieving these goals.

With this experience, I founded my own freelance company to help various laboratories improve the quality of their software production. For the past year, I have been working with labs and startups in Belgium, France, and Switzerland across different scientific domains, including neuroscience, MRI segmentation, and cell sequencing.


Session

07-18
13:55
30min
Design Patterns: Build better software from day one
Marine Guyot

Starting a new software project is exciting! It’s your chance to build something clean, functional, and easy to manage. But without a strong foundation, even small projects can quickly spiral into a confusing mess of bugs, untracked changes, and hard-to-follow code. This is especially true in research environments, where programming often takes a backseat to scientific discovery, and software evolves unpredictably as it adapts to new experiments and shared use. So how do you set yourself up for success from the beginning?

This talk is your beginner-friendly guide to designing smarter, maintainable software using proven design patterns. We’ll explore how patterns like the Singleton (for managing shared resources), the Template Method (for reusable workflows), and the Factory Method (for flexible object creation) can transform your code from fragile and frustrating to solid and scalable.

Using real-world examples from research labs, we’ll walk through how these patterns solve common problems, simplify collaboration, and keep your projects on track, even as they grow. Whether you're developing a new pipeline, a data analysis tool, or an experiment software, this talk will equip you with the foundational knowledge to design software that supports your scientific goals without becoming an unmanageable burden.

Let’s start building research software the right way, so you can focus on science, not debugging.

~ None of these topics
Terrace 2A