You can see what other projects Pierre is part of over on Pierre’s GitHub Profile.
Now let’s spend some time getting to know Pierre better!
Can you tell us a little about yourself (hobbies, education, etc):
The first code I wrote was an Applesoft BASIC program, on an Apple //e computer… I was 10 years old. Since then I always managed to bring computers in everything I did, at home or at work. As I was an amateur astronomer and was also very fond of physics in general, I chose to follow scientific studies. A few years later, I specialized in optics and photonics and graduated from Institut d’Optique Graduate School, which is now part of Université Paris-Saclay. I then pursued a PhD in the field of femtosecond lasers. Although it was mainly experimental physics, I had the opportunity to develop a code for simulating regenerative amplification in ultra-short pulse lasers; I learned recently that this code is still used today! After my PhD, I worked as a research engineer at THALES Avionics (on developing innovative head-up displays for aircrafts).Then, in 2007, I joined the French Alternative Energies and Atomic Energy Commission (CEA) where I was hired as leading software developer for applications involving image and signal processing as well as scientific instruments control. In 2012, I was given a project management position for the Laser Mégajoule timing and fiducial system development. Four years later, I was appointed head of a research laboratory. Lastly, in 2018 I had the opportunity to join Codra, an industrial software company, as a Project Director. In addition to this position, I am currently the pre-sales manager for the department of engineering at Codra. And of course, I’m also involved in open-source software development since 2008.
Why did you start using Python?
I started using Python in 2008, after a long and meticulous evaluation of various solutions that may fit my needs. Since early 2007 I was part of a research team at CEA. When I joined this team in 2007, every processing and acquisition software was written using commercial software. Some applications were getting huge and complex with a lot of GUIs for editing tons of parameters or visualizing results. Robustness was the main concern, therefore I chose Python since it was providing all the necessary tools for our research work (interactive computing and scientific data plotting) as well as the general-purpose libraries for building stable and robust applications. In 2008, when I started using and promoting Python amongst my colleagues, a piece of the puzzle was still missing: Python had no scientific-oriented IDE! That’s why during my vacations I began coding some tools for filling gaps in Python ecosystem, using Qt GUIs. After writing a variable explorer GUI that could be used directly from a Python interpreter to interact with current namespace, I wrote a Qt-based Python code editor, then a Qt-based Python console… and so on. After a few weeks only, this was done! This ultimately resulted in Spyder (Scientific PYthon Development EnviRonment), a scientific Python IDE that I first released to the public in September 2009: Python was finally a viable alternative to scientific commercial software. Today, thanks to a development team lead by Carlos Cordoba since 2012, Spyder is widely used for data processing and visualization with Python (est. 500,000 downloads/day).
What other programming languages do you know and which is your favorite?
What projects are you working on now?
At Codra, I’m involved in a lot of projects as a Project Director (or technical expert), in various fields like supervisory systems, data acquisition, multi-protocol gateways, data processing, data visualization, etc. From time to time, I even play the role of Project Manager. This is how I’ve been involved lately in CodraFT development, which was supported by CEA. It is available freely on GitHub: this is a Python-Qt based software aiming at processing signals and images and visualizing them. Its main upside is testability: the objective was to create a data processing software with a high level of robustness. Data processing features are mainly based on NumPy, SciPy and scikit-image.
Which Python libraries are your favorite (core or 3rd party)?
At the moment, I’m quite fond of scikit-image for image processing ; nice and clean API, and great documentation. OpenCV is also a great tool available to Python users and provides very efficient pattern detection algorithms for example.
What are some of the big lessons you learned while working on Spyder or WinPython?
I think that the most important lesson I’ve learned during those years is that we need to collaborate with other people. Otherwise, in the end, projects will at best remain as good ideas, or will be discontinued. With Spyder and WinPython, the thing that I’m the most proud of is that I managed to trust someone else to take over the projects and maintain them: in both cases, it was a good decision and projects are still active and popular.
Is there anything else you’d like to say?
I recently add the opportunity to attend a conference around Jupyter (PyData Paris). I really admire the work that has been done around the Jupyter ecosystem. From the IPython version I played with in 2008 to today’s JupyterLab, what an achievement from a technical point of view as well as in terms of community and project management!
Thanks for doing the interview, Pierre!