FOSDEM Quantum Computing Devroom
Date:
Excited to announce that I will be giving a talk at the Quantum Computing devroom @ FOSDEM 2025 organised by the Unitary Fund.
The talk is titled “No-one used my software: a tale of quantum software engineering”, and will talk about the practical experience of writing quantum software for a living over the years. We’ll touch on issues ranging from quantum physics and hardware to user interaction, requirements gathering and community building.
Abstract:
Quantum software engineering sounds complicated. Nevertheless, my experience is that the hard part of quantum software is not quantum, but software and engineering.
Paraphrasing a famous quote one might say that “quantum software engineering is a journey, not a destination”, a mantra that also fits this talk. As such rather than focusing on the destination I can divulge that the chapters of this story includes “How multiple inheritance killed my SDK” and its sequel “The return of the SDK” which will outline some the development process that led to Pasqal’s latest open-source SDK Qadence, a and a recurring entr’acte: “the GPU strikes back”, wherein the lure of GPU accelerated compute frequently leads our hero astray.
Building on the experience of over 3.5 years building software across teams at the full-stack quantum company Pasqal the presenter intends to bring a high-level view with concrete examples and demystify quantum software in a way that is accessible to the wider open-source community yet interesting and informative for practitioners.
And the notes submitted along with the abstract to constitute the proposal:
The talk will be high-level with a humerus tone, but we’ll use concrete(code) examples from Pasqals open-source software (https://github.com/pasqal-io) to exemplify our points, and the interplay with the wider FOSS community.
Most central to the talk will be Qadence, and it’s transformation path from a wrapper on top of common frameworks such as Qiskit, Cirq, Yao.jl where multiple inheritance was the weapon of choice, to an intermediate stage where we tried to make everything as functional as possible, to the ongoing compilation efforts in Qadence 2.
In the process of getting from A to B in the development stage many internal tools and prototypes were developed, and made available internally for testing either as standalone packages or integrated. Throughout the talk we will focus on the end-users and speculate as to why most of these softwares weren’t used much, even though in many cases they were explicitly requested from users. I believe this in common issue in quantum software that will be of interest to the community.
Hope to see you there!