Bjarne Stroustrup draws a full house at CERN for “21st Century C++”
The CERN Main Auditorium was full on Tuesday, 14 July, as Prof. Bjarne Stroustrup, the designer and original implementer of C++, delivered the EP/IT Data Science Seminar “21st Century C++”.
Interest in the seminar quickly outgrew the room. With the Main Auditorium at full capacity, the Council Chamber was opened as an additional breakout space. Online participation was also strong, with 173 people connected via Zoom and 254 following the webcast. Overall, the seminar reached an estimated audience of around 900 people. The event was part of the CERN STEAM Academy Seminar Series, with the support of CERN’s Next Generation Triggers Project.
For many at CERN, C++ is not an abstract topic. It is part of the software backbone behind scientific computing, data processing, simulation, reconstruction and real-time systems. Prof. Stroustrup’s seminar was therefore not only a look at the evolution of a programming language, but also a timely reflection on the tools that help make large-scale science possible. In his talk, Prof. Stroustrup addressed a common misconception: that C++ is still best understood through the habits and limitations of its early decades. Instead, he presented modern C++ as a language that has continued to evolve, with strongly typed generic programming, modules, systematic error handling and safe resource management at its core.
The message was clear: 21st century C++ is not simply about writing faster code. It is about writing code that can be smaller, safer, more maintainable and more reliable, while still meeting the performance demands of modern computing. This is particularly relevant in environments such as CERN, where software must often operate at the intersection of scale, precision and speed. As experiments prepare for ever larger data volumes and increasingly complex computing challenges, the way scientific software is designed, written and maintained becomes a central part of the research process.
The strong turnout also marked an important moment for the CERN STEAM Academy, now in its first edition. The Academy brings together participants from diverse backgrounds for a 10-week programme at CERN focused on software technologies, edge computing, data science and machine learning. Its seminar series opens these discussions to the wider CERN community, connecting the Academy with leading voices in computing, AI and science.
The CERN STEAM Academy Seminar Series continues on 28 July 2026 with Prof. Maciej Besta from ETH Zurich, who will give the seminar “Towards Effective and Efficient Reasoning Language Models”.
Here you can find all the info on the upcoming seminars: https://indico.cern.ch/event/1697464/page/43475-steam-seminars
The full recording of Prof. Stroustrup’s seminar is available on CDS Videos:
https://videos.cern.ch/record/3026440
Photos from the event are available on the CERN Document Server:
https://cds.cern.ch/record/2966572