CppCon 2026

Floating point has a reputation for betrayal. Change the thread count, vector width, compiler flags, reduction tree, or target architecture, and the low bits can move. Parallel algorithms make this worse: the standard often specifies the operation, but not the numerical expression whose result must be reproduced. This talk asks a provocative question: what if reproducible numerics did not have to be slow?

We will show reproducible, deterministic implementations of reduce and scan that exhibit better error behavior on hostile floating-point workloads and can match or beat conventional standard-library implementations on realistic workloads. The trick is not to freeze the execution schedule. It is to specify the expression being computed, then let the implementation use SIMD, threading, blocking, tiling, and platform-specific strategies to compute that expression efficiently.

The key idea, developed through C++ standardization work such as P4016R0 and P4229R0, is reproducibility by reproducing the computation. Instead of asking the implementation to promise a particular schedule, we give the calculation a named expression. Once that expression is chosen, changing the thread count, vector width, chunking, or blocking strategy does not silently change the answer.

A reproducible scan makes this harder than reduce because it does not expose only one final value. It exposes every prefix. A reproducible final sum is not enough if the intermediate results still drift. We will show how expression and observation contracts make those prefixes reproducible without forcing the computation back into a slow sequential order.

Then we go below the algorithm layer, to the places where bits usually escape: FMA contraction, denormals, floating-point environment choices, math-library approximations, and vectorized transcendental functions. The goal is not to get the same answer by turning off the hardware. We will show reproducible vectorized primitives, including transcendental functions, running at speeds comparable to established vector math libraries while preserving a cross-platform numerical contract.

Finally, we put the whole stack under stress: a heterogeneous numerical pipeline across x86-64, Apple Silicon, and CUDA. The data is deliberately hostile, with high cancellation rates and fragile intermediate states. The aim is not to pass friendly benchmark cases, but to reproduce the specified computation, including the same intermediate failures, not just the same final answer, bit for bit, across CPUs, GPUs, and toolchains.

An overview of contemporary hardware platforms and how software engineering and design practices and methodologies could help in building performant systems, with a particular focus on low level optimizations. Contrary to attempting direct low level software engineering for addressing performance specifics, this talk would focus on how higher level abstractions could and should be used, and how a software engineer could help the compiler to “do the right thing”. The trend of moving software engineering focus upward to constructs that have a tendency of hiding the specific of the underlying platform is quite clear – and there certainly are very good reasons for such a paradigm change.

Performance matters. Premature optimization is evil. Is there anything in between those two extremes? How feasible is it to rely on the abstraction of the platform as hidden by the compiler and the corresponding libraries and language constructs, expecting that it will be able to realize the intended lower level optimizations? How much of hints and specifics would one need to expose for the compiler to be able to get the equivalent of direct low level approach? What is the right balance between expressing application logic via higher level construct yet still being able to gain the performance benefits compared relying on the low level specifics?

Looking from the practical applicability of lambdas, iterators, ranges, coroutines, error handling, and safe(r) memory access, the talk would attempt to cover a set of use cases with a focus on analysis of what could be done for focusing on performance.

The overall goal of the talk is not to go deep into low level aspects; the goal is to explain that one needs to be aware of such low level details while operating on the higher level constructs.

Memory leaks are very hard to treat. We have reliable tools like valgrind or AddressSanitizer to tell us whether or not an application has leaked memory, but current tools cannot tell us where the leak happens (they can tell us where the allocation happened,which is not the same).

I show that it is possible to run an analysis that is equivalent to running a garbage detection pass after every instruction and use that to find the actual cause of memory leaks in open source applications.

Using a garbage detection algorithm on record-and-replay recording, one can overcome the most obvious obstacle: stopping a program a program after every instruction or running garbage detection algorithms is very slow, but we can effectively run a bisection search on the recorded timeline to avoid having to evaluate after every instruction.

The goal is attribution to source code. Are we limited to reporting the instruction that overwrites the last reference to allocated memory or can we actually diagnose a specific error for a concrete function?

Building embedded systems wastes time on infrastructure instead of features. Before running application logic, developers lose hours bootstrapping init systems, wiring services, debugging startup failures, and fighting tooling never designed for constrained or early-boot environments. AEMBER is a developer-first PID1 (init system) that eliminates this overhead by providing a modern C++ runtime for process supervision, container orchestration, and service management - letting you focus on your application, not your plumbing.

This talk demonstrates how C++23 enables robust embedded systems without sacrificing performance. We'll explore std::expected for exception-free error handling, if consteval for compile-time optimization paths, and deducing this for zero-overhead policy classes. You'll see how monadic operations compose system calls into clean pipelines, and how modern C++ features build type-safe APIs for namespaces, cgroups, and process management.

Starting from main(), we'll trace AEMBER's architecture: how components compose, how errors propagate through std::expected chains, and how C++23 patterns enable embedded systems to be both safe and fast. We'll wrap up with a live demo showing AEMBER managing containers and services in real-time. You'll leave with concrete techniques for building maintainable embedded infrastructure using cutting-edge C++.

Modern C++ development spans platforms, build systems, and increasingly complex project structures. In this landscape, Visual Studio Code has carved out its distinct position as a lightweight, extensible editor that works where you do, with first-class extensions built on decades of C++ language services and build system integration. This year's update deepens both of those foundations and opens them up to AI-driven workflows powered by GitHub Copilot.

This session walks through what's new across three dimensions of C++ development in Visual Studio Code. For the inner loop, we'll cover practical improvements to everyday compile-test workflows, including CMake target bookmarks and Run Without Debugging. For writing and understanding code, we'll show how the C/C++ extension's language services now surface rich project understanding to GitHub Copilot, giving AI assistance real awareness of your code structure instead of relying on generic pattern matching. For your workflow, we'll demonstrate how AI capabilities compose across different modes of working: inline suggestions for hands-on editing, agent coordination for multi-file refactoring, and CLI sessions for long-running tasks and deeper codebase analysis you can revisit on your own schedule.

Throughout the session, we’ll use real-world C++ problems to demonstrate where these tools meaningfully improve productivity and where they intentionally stay out of your way. Finally, we’ll show you how to inform GitHub Copilot about your team's best practices and scale them across your code through extensibility points like skills, custom instructions, and custom agents.

Modern storage hardware has evolved at a breakneck pace. PCIe Gen 5 NVMe drives can push data at 10 GB/s, yet standard C++ file abstractions often leave them starving for data. Many developers think they need to abandon portability in favor of unmaintainable, OS-specific kernel bypasses to achieve the throughput necessary to saturate these drives. This session is for developers writing high-throughput, performance-critical applications who want to hit bare-metal speeds without sacrificing clean architecture, cross-platform support, or type safety.

Using a high-throughput cryptographic hashing engine as a concrete case study, this presentation demonstrates how to design a hardware-saturating data pipeline built entirely on the idioms of C++26. We will explore how the mathematical design of an algorithm - specifically BLAKE3's binary Merkle tree structure - can be mapped directly to wide SIMD vector lanes and concurrent CPU cores. We will walk through an execution model that completely bypasses the OS page cache, orchestrates memory without heap allocations on the hot path, and unifies OS kernel quirks in a portable way.

By the end of this presentation, you will learn how to replace rigid thread pools with lock-free, asynchronous execution graphs using Sender/Receiver paradigms (std::execution) and vectorization (std::simd). Crucially, we will focus on the build engineering required to make this work today. We will cover how to use advanced CMake tooling to safely compile multi-architecture vector binaries from a single source of truth, how to prevent LTO cross-contamination, and how to structure your pipeline today to seamlessly absorb upcoming C++ features in a world of trailing vendor toolchains.

In many languages, you can mock almost anything. Python has unittest.mock, Java has Mockito and PowerMock, and C# has Moq. In C++, you can usually mock only what was designed to be mockable — and little else. Want to mock a free function? Wrap it in an interface. A static method? Refactor to a template. A non-virtual member? Redesign the class hierarchy. The cost is real: developers either shape production code around testing-tool constraints instead of domain needs, or they simply leave hard-to-mock code untested.

This talk introduces [LibraryName], a new C++ mocking library we developed (and plan to publish soon as open source) that can mock almost any C++ callable — free functions, static methods, non-virtual members, private members, templates, lambdas, and even extern "C" routines — without modifying the code under test. Developed and battle-tested at [CompanyName] for three years across a very large C++/C codebase, it aims to deliver a first-class testing experience for C++. Under the hood it uses runtime binary patching — replacing function prologues with redirections at test time, restoring them on scope exit — all surfaced through familiar Google Mock syntax.

We'll show [LibraryName] in action, walk through the patching mechanism, and share how to get started -- you'll walk away ready to test almost any function, regardless of how it was designed. Strong test coverage has always been essential, and AI-assisted code generation only increases the need for reliable testing backpressure. C++ deserves mocking tooling that matches.

Clang is the C++ compiler used by hundreds of millions of developers every day, yet most of those developers have never looked inside it. The codebase can feel intimidating, with millions of lines of C++, an unfamiliar architecture, and a review process with its own customs. But once you know the map, contributing is surprisingly accessible.

This session is a live, code-first walkthrough of the full contribution cycle for Clang. Starting from a clone of the LLVM monorepo, we will configure a development build, locate the right subsystem for a real reported bug, write a regression test, fix the bug, and verify the fix. We will then extend that foundation to implement a small but complete language feature: adding a new diagnostic, walking through Sema, the AST, and the diagnostic engine as we go. Every step attendees see on screen can be replicated on their own laptops during and after the session.

Attendees will leave with a mental model of Clang's layered architecture (driver, frontend, Sema, CodeGen), a workflow for finding the code responsible for any given compiler behavior, and enough familiarity with the test infrastructure and review process to open their first pull request.

CUDA Tile C++ is a novel C++ DSL and compiler that automatically parallelizes array operations across GPU threads while presenting a single logical thread to the programmer. In CUDA Tile C++, parallelism is expressed through SIMD-style operations on arrays of data called "tiles" while memory accesses are written using structured memory abstractions. The compiler analyzes the tile and memory operations to generate an equivalent multi-threaded program that leverages the advanced hardware capabilities of modern GPUs. This allows the user to write simple programs that exploit GPU parallelism without needing to manage low level hardware concurrency.

As GPU architectures evolve to meet the demands of high performance computing, the traditional multi-threaded GPU programming model becomes increasingly unwieldy. To achieve optimal performance, the GPU programmer must manage a growing variety of hardware features including shared memory, Tensor Memory Accelerators, and Tensor Cores. Traditional CUDA code is deeply coupled to the underlying hardware resulting in complex and non-portable algorithms. The CUDA Tile C++ compiler removes the need for coordinating these low level hardware details, freeing the user to reason about their algorithm rather than asynchronous code.

The tile programming model introduces three key abstractions:

• cuda::tiles::tile - a multi-dimensional array with value semantics whose operations are parallelized across the GPU
• cuda::tiles::tensor_span - a view into a multi-dimensional in-memory array with an interface similar to std::mdspan
• Tile Views - adapters that present a uniform "chunking" of a tensor_span which gives the compiler visibility into memory access patterns

In this talk, we will explore how these abstractions allow NVIDIA's Tile C++ compiler to schedule the program across hardware threads all while preserving the illusion of a single-threaded abstract machine to the user.

AI agents are accelerating code generation at an unprecedented pace, yet C++ file is still compiled as a monolithic translation unit, meaning a many-core machine often cannot rebuild a template-heavy file much faster than a single core. Header-heavy designs, inline functions, and templates compound the problem: the same bodies are parsed repeatedly, and small edits can invalidate an entire translation unit.

This talk presents a compiler-agnostic technique for splitting translation units into smaller independently compilable units. Using libclang, the tool rewrites headers and sources into declaration-only interfaces plus generated .cpp fragments, backed by a shared preamble that can turn into a C++ module, precompiled header, cached serialized AST and a source-map. The fragments compile in parallel, then link into a binary functionally equivalent to the original monolithic build.

Expect benchmarks from major codebases covering compile-time, link-time, and run-time performance; live demos of fast incremental updates that avoid reprocessing unchanged code as well as an analysis of the resulting binaries layout.

Modern C++ development has embraced package managers like Conan and vcpkg for dependency management, gaining reproducibility and productivity. However, a significant weak point remains: the build toolchain. Unlike standard dependencies, the disparate set of tools comprising the compiler, linker, standard libraries, and OS-specific headers often remains outside this managed ecosystem. This gap leads to frequent complications, where misaligned versions between local environments and CI cause build errors and subtle ABI incompatibilities.

The operational cost of maintaining these toolchains is high. Updating a compiler version often necessitates the creation and deployment of new VM images or manual infrastructure updates, a process that can be even more complex if specialized environments like CUDA are involved. These hurdles directly impact engineering productivity and organizational alignment. Docker is a popular solution to isolate toolchains, but it does not support all the mainstream platforms.

This talk proposes a shift in perspective: treating the build toolchain as just another dependency. We will explore how leveraging toolchain isolation allows developers to express specific toolchain versions within their project manifests—using cross-platform tools like Conan and vcpkg to ensure strict repeatability across local developer environments and CI machines.By automating the fetching and configuration of the entire compiler suite, teams can ensure perfect alignment across all environments, simplify the upgrade path, and ultimately boost engineering velocity.

The C++ Core Guidelines is quickly gaining popularity as the main C++ coding standard in the industry. The official introduction of the standard states the following: "Many of the rules are designed to be supported by an analysis tool. One way of thinking about these guidelines is as a specification for tools that happens to be readable by humans."

We took up this challenge: we built the tools, implemented a subset of the C++ Core Guidelines, and applied them to more than 8,000 industrial embedded software projects. That was not an easy undertaking, because the defined “rule enforcements” contain all kinds of caveats, shortcomings, unclarities, exceptions etc.

During this talk, we will explore the conceptual obstacles we encountered during implementation, how we have addressed them, and what remains to be done to make the C++ Core Guidelines the de facto standard in C++ programming.

Every Boost library that graduated to the C++ Standard became invisible. shared_ptr doesn't say "Boost" anywhere in . Boost's greatest successes carry no signature — and that invisibility is killing recruitment. This talk is not a status report. It is an answer to the question: why would I give my best work to a project that erases my name?

We begin with the people. Peter Dimov. Beman Dawes. Howard Hinnant. Marshall Clow. These engineers didn't just write libraries — they defined what production-quality C++ means. We will say their names, show their faces, and then ask the only question that matters for Boost's future: who replaces them? Every technical detail in this talk serves that question.

The answer starts with the peer review process — not as a quality gate, but as an apprenticeship. Boost review doesn't just catch bugs; it forges people. It is the mechanism by which a competent developer becomes a Dimov-class designer, and we will show how that pipeline works today.

Then the technical core. Boost.Unordered's unordered flat map — not as a benchmark, but as a character study. Joaquín López Muñoz sat down, studied SIMD probing, and built something faster than a team at Google. We won't just show a bar chart; we'll show the design decisions — the one that made it fast and the one that almost kept it from shipping. Boost.Intrusive — presented as a magic trick before an explanation. One object. Three containers. Zero allocations. We'll let the audience sit in "wait, that can't work" for ten seconds before we reveal the mechanism. Boost.Decimal — IEEE 754 decimal floating-point that feels like a native type, with portable performance rivaling GCC's non-portable _Decimal extensions. We will also cover Boost.JSON, Boost.URL, Boost.Scope, Boost.Cobalt, and the design thinking behind each.

Networking gets its own act. Beast, Corosio, Capy — presented as a list, they're alphabet soup. Presented as what they are, they become a statement of intent: Boost is building the networking layer the committee has been unable to standardize for fifteen years, and we believe we can ship it. We will show how buffer sequences, stream concepts, and type-erased I/O deliver ABI-stable networking compiled once against an abstract transport, and how this connects to active WG21 proposals for C++29.

We close not with a mailing list URL but with a name. One person — a contributor who submitted their first Boost PR this year, went through review, and is now preparing a library for formal consideration. A person is a door; a URL is a wall. The thesis of this talk is not "Boost is alive." The thesis is that Boost is where you become the engineer you want to be — and every section, from the review process to the mentorship lineage to the fact that one person can build the fastest hash map in the industry, is proof.

For almost 30 years, Visual Studio has been a core part of the C++ developer's toolkit on Windows. This year, we're building on that foundation with investments across three themes: making the IDE faster and more responsive for large codebases, advancing compiler conformance and runtime performance, and integrating AI-powered workflows that help you debug, refactor, and optimize more effectively. Beyond the IDE, we'll show you how AI-powered command-line tools can become a natural part of your development process. This session combines demos and practical guidance so you can get the most out of these tools right away. Come with curiosity, leave with new techniques you can apply immediately.

Notifiers are a critical synchronization primitive in task-parallel programming systems such as Intel TBB and Taskflow, responsible for efficiently sleeping and waking worker threads as tasks become unavailable and available over and over again, directly impacting scheduler throughput and latency. Correctness here is non-negotiable: a single missed wakeup can significantly hamper the performance of an entire program. Yet writing strong unit tests for notifiers is notoriously difficult, because the bugs they target, lost wakeups, spurious wakes, race conditions, are timing-dependent, non-deterministic, and often only surface under specific thread interleavings that are hard to force reliably.

The problem is compounded in practice. Notifier implementations evolve constantly: small algorithmic tweaks, memory ordering changes, and refactors across systems demand a fresh round of carefully constructed tests. This is tedious, expertise-heavy work that takes a lot of time and engineering effort. In this talk, we explore using Large Language Models (LLMs) to automate the generation of the unit tests for notifiers. Specifically, we will demonstrate how LLM-generated tests, guided by proper prompts can systematically stress the two-phase wait protocol across Notifiers in Taskflow. We will show this in a widely used Notifier implemented in Taskflow. We are able to find an undiscovered bug that has been existing in the project.

The C++ standard is moving more and more in the direction of safety. In the process, the committee is providing us with tools to help make our own code safer.

Many of these tools are specifically around numerics and none of them are enabled by default! To make matters worse, most C++ programmers don't even know these library features exist! Even if they did know, they wouldn't use them, because they are far too wordy.

We're going to do a survey of the relatively recently added new numeric safety related language features and see how they might come together in a real-ish project.

Language bindings for C++ can be valuable tools for developers who wish to make C++ libraries available for use in other programming languages, like Python. There are many reasons developers may wish to do this, including testing, prototyping, or performance.

One of the issues often faced when writing bindings for C++ is the need to write excessive amounts of boilerplate code to bridge the gap between the C++ types and interfaces and their equivalents in other languages.

Building on previous presentations, this talk will not just explore how you can use Reflection in C++26 to generate the necessary binding code, but also how you can integrate this into a CMake-based build system to enable the automatic generation of bindings for existing classes or libraries with minimal additions. We will present real-world examples, primarily using Python, to show how language bindings can be automatically generated, but we also show how this can be extended to other languages.

First day on a new project. You set up your workstation and then download a repository of over 1,000,000 lines of code. Even more intimidating, the code has been around for 20 years. Parts are in legacy C++, parts use the latest C++26 features, and increasingly, AI coding assistants are weaving new code through all of it, meaning some parts of the codebase may have never been fully understood by a developer. You feel overwhelmed. Don't Panic! In this talk, we share techniques and tools that real engineers use to find their bearings in code they didn't write, from the humble grep through debuggers and profilers to time-travel debugging and AI-assisted exploration. The audience will leave this talk with a mix of simple and advanced tricks for navigating large and complicated codebases.

Most bosses aren’t stupid, they just seem like it sometimes! This can be especially so when a great programmer first becomes a manager – suddenly the very attributes that made them good at their job become a problem. Whether you’ve just been ‘promoted’ to a management position, you hope to be some day, or you want to know how to deal with a difficult manager – it’s very useful to understand the world from a boss’s perspective. Not all leaders are bosses (and not all bosses are truly leaders): if you want to get something meaningful done then you’ll need to influence other people.

Whether or not you're a boss, leadership is important. To do meaningful/large things, we need to work with other people. Encouraging and persuading peers in the right way will make your and your peers' working lives better.

Lots of war stories (anonymised to protect the guilty) and tips and tricks, this talk is basically 'how to win friends and influence people' for programmers.