Marco Arena

Founder of the Italian C++ Community

---

Topic: Welcome and Closing Messages

Marco is the founder of the Italian C++ Community. He works at Tetra Pak as one of the technical leaders of the AI Engineering Team, pioneering the development of AI-based applications for packaging lines. From 2011 to 2019, he served the Ferrari F1 Racing Team (Scuderia Ferrari) as a Software Engineer and C++ Specialist. Marco is very active in the tech communities environment: he founded the Italian C++ Community in 2013, Coding Gym in 2016, ML Modena in 2020, and he collaborates with ConoscereLinux. Since 2016, Marco has received the Microsoft MVP award for his commitment to divulgation.

Close

Massimo Fioravanti

Compiler researcher at Politecnico di Milano

---

Topic: Interactive Program Design in C++: A Taxonomy for Practitioners

Massimo Fioravanti is a C++ programmer who has long suspected that something was fundamentally flawed in how interactive programs are designed. Now a compiler researcher at Politecnico di Milano, he's working to fix that. Massimo is the main maintainer of Rulebook, a collection of utilities aimed at asymptotically reducing the complexity of building interactive systems in mainstream languages like C++ and Python.

Close

Marco Marcello

Team Lead at Bloomberg

---

Topic: 8 Queens at Compile Time: A Journey from C++03 to Modern C++

Marco Marcello is a Team Lead in the Real-Time Data Engineering group in Bloomberg’s Frankfurt Office. He has led a team responsible for rule-based processing systems since 2023. His interests include modern C++, concurrency, and the evolution of metaprogramming.
He has held various engineering roles across Bloomberg’s offices in London, Switzerland, and Germany, focusing on order management systems, financial analytics, and high-performance, data-driven applications.He previously worked in the UK on mobile technology at Samsung Electronics, and at ARM Ltd., working on JTAG emulation of ARM test chips.
Born in Cagliari, Sardinia, Marco began programming on the iconic Commodore 64. He earned a master’s degree in electronics engineering from the University of Cagliari in 2001.

Close

Jonathan Marriott

Software Engineer at Bloomberg

---

Topic: 8 Queens at Compile Time: A Journey from C++03 to Modern C++

Jonathan Marriott is a Software Engineer at Bloomberg, where he works within the Performance Analytics & Data Engineering group. Jonathan specializes in C++ development, working on a number of high performance systems. He earned a master's degree in mathematics and computer science from the University of Bristol.

Close

Denis Yaroshevskiy

Performance Engineer

---

Topic: SIMD substring in a string

Denis works at Meta doing performance for a living.
Denis is a semi-active C++ community member, a few talks, one paper (failed), some C++ standard library patches, some Chrome base library contributions. Denis is one of the maintainers for EVE opensource library (SIMD in C++20).

Close

Alberto Barbati

Senior R&D Video Game Developer

---

Topic: Zero or More: A Tour Through C++ Techniques for Returning Multiple Values

Alberto Barbati is a passionate C++ programmer who has been active since the early days of the language and has been following the work of the C++ standards committee since 2008. Since 2001, he has specialized in video game programming, contributing his expertise at Ubisoft Milan, Reply Game Studios, and currently at Nacon Studio Milan. Alberto has also shared his knowledge as a Core Trainer in Game Programming at the Digital Bros Game Academy.

Close

Stefano Fiorentino

Head of CC IoT & Embedded Systems at adesso Schweiz AG

---

Topic: Mocking the UART in C++: A Hands-On Approach

Stefano is a professional engineer in Telecommunication Engineering. He started developing software in MATLAB and C for a ray-tracing and earth subsurface tomography, a cluster computing solution. Then he moved working with Qt and C++ for a GUI-based subsurface imaging product. After a really nice period with plain Modern C++ in a financial cluster computing institution, he encountered the Yocto Project and never left it. He is currently Head of the Competence Center Embedded Systems at the consulting firm adesso Schweiz AG. Father of two daughters, high-handicapper golf player, host at C++ User Group Lugano and ISO C++ Committee Member, co-founder of Italian Embedded.

Close

Daniele Pallastrelli

Senior Software Engineer

---

Topic: Delegating Behaviors in C++: A Practical Tour of the Available Mechanisms

Daniele Pallastrelli has been programming and designing software for nearly 30 years, ever since he discovered his passion for it as a teenager in the mid-90s.
He is a professional software engineer, speaker, and author — and an enthusiastic runner.
Although usually reluctant to talk about himself in the third person, he can occasionally be persuaded to do so.
In his spare time, Daniele enjoys writing technical papers and blog posts,sharing insights from his long experience with the C++ community.

Close

Zero or More: A Tour Through C++ Techniques for Returning Multiple Values

07:45 - 08:35

---

From raw buffers to generators, the journey of returning an arbitrary number of elements in C++ reflects the evolution of the language itself. C++98 relied on C-style pre-allocated buffers, though output iterators provided an alternative. C++11 made returning containers efficient with move semantics, and lambdas introduced allocation-free flexibility. Then came C++20 with `span` and ranges, unlocking composability — and now C++23 introduces generators, making it all feel effortless.
Each technique has trade-offs (yes, even buffers have their place!). This talk walks through the strengths, pitfalls, and use cases for each approach — because knowing all the tools in your toolbox makes you a better programmer.

Speaker

---

Close

Delegating Behaviors in C++: A Practical Tour of the Available Mechanisms

08:40 - 09:30

---

Delegating behaviors to other parts of the code is essential for writing reusable, extensible, and flexible software.
C++ provides several mechanisms to achieve this: from virtual functions to function objects, from function pointers to lambdas, as well as templates, concepts, and more.
Each comes with its own strengths, trade-offs, and sweet spots.
In this talk, we will systematically explore the main options the language offers for passing behaviors — comparing their syntax, performance, flexibility, and suitability for different contexts.
The goal is to provide a reasoned roadmap that helps developers consciously choose the most appropriate mechanism based on their specific needs.

Speaker

---

Close

SIMD substring in a string

10:00 - 11:00

---

Despite being widely avaliable on modern hardware, vectorization remains a dark art known to only a select few. For the last few years I've been trying to change that by explaining (with pretty pictures) how to do different simd algorithms.
Searching a subsequence in a sequence is next one to tackle (you don't need to watch anything beforehand).
We will start with the "interview problem" level solutions and examine how they perform (spoiler: poorly).
Then we look at the standard libary and see how it compares to the interview solutions (spoiler: the standard library should do a better job).
Finally we are going to do some SIMD, using EVE library.
We'll do:

• memcmp
• memchr
• and finally a eve::algo::search, which is a pretty good substring in a string.

Speaker

---

Close

Mocking the UART in C++: A Hands-On Approach

12:00 - 12:35

---

In the world of software engineering, robust and efficient testing is crucial for ensuring reliable software. But what happens when your code interacts with hardware?
This session provides a step-by-step walkthrough of mocking UART interfaces in C++. Through a practical, demo-driven approach, we’ll explore how to decouple your code from hardware dependencies, simulate UART behavior, and build effective unit tests, even at compile time.
Best of all, we’ll achieve this without relying on any testing frameworks, just plain C++. By the end of this talk, you’ll have a clear understanding of how to integrate mocking techniques into your development workflow, enabling more reliable and maintainable embedded software.

Speaker

---

Close

Interactive Program Design in C++: A Taxonomy for Practitioners

12:40 - 13:30

---

Most C++ talks focus on algorithms or data structures that run once and exit. Yet, in real-world software much of our code lives inside interactive programs that converse with human users or remote services. Translating business-level interaction scenarios into efficient, maintainable C++ can be trivial, or fiendishly hard.
This session gives a concise formal map of every major class of interactive program and shows how to realize each in C++ with little or no runtime cost. Starting with the simplest sequential dialog, we climb the hierarchy of complexity. We will discuss tree-structured, graph-structured, push-down, and finally Turing-complete programs, and then layer commonly required features such as state serializability, pre-condition checking, and state inspectability.
For every stop you will receive:

• A motivating example of the problems found there
• A explanation of which cpp language features are required to implement it
• A “cheat-sheet” pattern (code + design) ready for production

Finally, we will see what limitations prevent idiomatic implementations are inherent to CPP language features, and what solutions are available to mend them.

Speaker

---

Close

8 Queens at Compile Time: A Journey from C++03 to Modern C++

14:00 - 14:50

---

The 8 Queens Puzzle — placing eight queens on a chessboard such that no one threatens another — is a classic computer science and combinatorics problem, with exactly 92 valid solutions. While traditionally solved using runtime algorithms, this talk will revisit the puzzle through the lens of compile-time computation, illustrating the evolution of C++ metaprogramming.
Years ago, one of the presenters implemented a solution using template metaprogramming techniques available at the time, demonstrating how even complex problems could be solved entirely at compile time. In this talk, we will revisit their original approach and contrast it with modern solutions that utilize variadic templates and constexpr functions to express the same logic more clearly and concisely.
Through this journey, we will highlight not only how compile-time capabilities have grown, but also how modern techniques can drastically improve readability, maintainability, and performance. This talk is for anyone interested in metaprogramming, compile-time computation, or simply in seeing how far C++ has come in enabling elegant solutions to classic problems.

Speakers

---

---

Close