Bryce Adelstein Lelbach
where else to find them
what they talk about
all 12 in the filters
where they keep going back
- CppCon8 editions2014–2021
- ACCU4 editions2019–2025
- Meeting C++3 editions2019–2024
all 9 in the filters
on a stage with
- Alisdair Meredithonce
- Andreas Fertigonce
- Andy Terrelonce
all 22 in the filters
who talks about the same things
- Herb Sutter31 talks
- Rainer Grimm26 talks
- Andreas Weis21 talks
- Kris Jusiak31 talks
- Timur Doumler45 talks
- Vittorio Romeo27 talks
| title | conference | date |
|---|---|---|
| Block-Based Parallel Programming | NDC TechTown 2025 | 2025-09-22 |
NVIDIA presents cuTile, a block (tile)-based kernel model compiled to Tile IR that JITs Python kernels (C++ API coming) and maps array-centric code to GPU… | ||
| GPU Programming in Pure Python | PyCon US 2025 | 2025-05-16 |
Claim is that you can write GPU kernels in pure Python and get performance parity with CUDA C++ using the CUDA Python object model plus Numba CUDA and the CUDA… | ||
| The CUDA C++ Developer Toolbox | ACCU 2025 | 2025-04-01 |
Shows concrete rules and code patterns: prefer Thrust and libcu++ over hand‑written kernels when they meet needs, fuse transform+reduce with transform_reduce… | ||
| The C++ Execution Model | ACCU 2025 | 2025-04-01 |
Reads the C++ standard and extracts concrete rules: a platform-agnostic flat global storage model, the as-if rule, and how threads evaluate function calls… | ||
| The C++ Execution Model | Meeting C++ 2024 | 2024-11-14 |
C++ defines a platform‑agnostic abstract machine with flat storage and threads whose evaluations are ordered by sequenced before, synchronizes with and… | ||
| Accelerating GPU Algorithms in pure Python | PyData NYC 2024 | 2024-11-06 |
Claims that pure Python kernels JIT‑compile to the same code and match CUDA C++ performance, demonstrated with a copy, tiled matrix transpose, and batched… | ||
| C++ Horizons | NDC TechTown 2024 | 2024-09-09 |
Claims three concrete C++ changes will reshape code: reflection/injection that extracts metaobjects, builds token sequences, queues injections and… | ||
| Think Parallel: Scans | CppNorth 2024 | 2024-07-21 |
Shows how the inclusive scan becomes a parallel primitive using a decoupled lookback protocol and a scan tile state object that replaces O(input) temporaries… | ||
| Think Parallel | ACCU 2024 | 2024-04-17 |
Argues that a parallel inclusive-scan primitive, implemented with decoupled look-back, replaces the classic two-pass upsweep/down-sweep and enables… | ||
| C++ Horizons | Meeting C++ 2023 | 2023-11-12 |
Concrete demos show compile-time reflection/injection that uses the rifi/reflection handle, splice and identifier splice with a template 4 to generate enum… | ||
| Endnote: AI-Assisted Software Engineering | C++ on Sea 2023 | 2023-06-28 |
Claims that AI excels at analysis tasks and struggles with creative code generation, demonstrated by a prototype CY diagnostic tool on GitHub that identifies… | ||
| Keynote: C++ Horizons | ACCU 2023 | 2023-04-19 |
C++ is set to add reflection-based enum name extraction, compile-time injection for generating wrappers and struct-of-arrays transformations, and hash_append… | ||
| C++ Standard Parallelism | CppNorth 2022 | 2022-07-18 |
Standard C++ now offers a portable parallelism model that claims to run on CPUs, GPUs and clusters without vendor extensions, via three pillars: a corpus of… | ||
| Multidimensional C++ | CppNorth 2022 | 2022-07-18 |
C++23 introduces mdspan, a non-owning multi-dimensional view with parameterized layouts, extents and slicing, but it intentionally omits iterators because… | ||
| C++ Standard Parallelism | C++ on Sea 2022 | 2022-07-05 |
Standard C++ already provides portable parallelism: transform_reduce word count shows a single parallel algorithm mapping to CPUs or GPUs; execution policies… | ||
| C++ Standard Parallelism | CppCon 2021 | 2021-10-24 |
Roadmap claims: standard C++ gains portable parallelism via concrete pieces — execution policies let parallel algorithms dispatch to vendor libraries… | ||
| C++ Committee's Library Evolution Working Group | CppCon 2020 | 2020-09-14 |
Discussion lays out concrete progress and process: executors aim for C++23 but need more review, the incubator and mailing-list reviews speed papers, and the… | ||
| [old version]The C++20 Synchrononization Library | Meeting C++ 2019 | 2019-11-14 |
C++20 adds concrete primitives and a working demo: a small tasking runtime implements a bounded MPMC queue, jthread with stop_token and proper… | ||
| The C++20 Synchronization Library | Meeting C++ 2019 | 2019-11-14 |
Claims a compact, runnable tasking runtime that actually builds: a bounded MPMC queue with two semaphores, a thread group using J thread with stop_token… | ||
| Modules are coming | Meeting C++ 2019 | 2019-11-14 |
Claim: C++20 modules redefine the compilation model, replacing textual inclusion with precompiled module interfaces (BMIs) that cut redundant parsing, enforce… | ||
| PREVIEW VIDEO - November | ACCU Autumn 2019 Belfast | 2019-11-11 |
Two-day ACCU Autumn conference runs 2019-11-11 to 2019-11-12 at the Hilton in Belfast and deliberately abuts the WG21 C++ meeting in the UK so attendees can… | ||
| The C++20 Synchronization Library | CppCon 2019 | 2019-09-15 |
Summarizes the C++20 sync additions and shows a working bounded-depth tasking runtime: uses std::atomic::wait/notify, std::atomic_ref, counting_semaphore… | ||
| “The C++ Execution Model” | CppCon 2018 | 2018-09-23 |
C++ defines an abstract machine of threads evaluating code in a flat storage space, not modeling caches or NUMA; memory_order_consume is ignored here. Rules… | ||
| C++ Execution Model In Five Minutes | C++Now 2018 | 2018-05-06 |
Defines expressions as value computations plus initiation of side effects and separates value computations from side-effects; side effects can be visible… | ||
| C++17 Features (part 1 of 2) | CppCon 2017 | 2017-09-24 |
C++17 consolidates practical language changes: structured bindings let auto [x,y] = point3 to name members (and work with tuples/arrays); selection statements… | ||
| C++17 Features (part 2 of 2) | CppCon 2017 | 2017-09-24 |
C++17 adds concrete library features: string_view: non-owning pointer+length avoids copies, but becomes invalid if the source moves; std::optional: no-heap… | ||
| Multidimensional Iterators | C++Now 2017 | 2017-05-15 |
Attempts to abstract nested loops into reusable iterators usually kill vectorization, but here a generate_indices range implemented with C++ coroutines emits… | ||
| “The C++17 Parallel Algorithms Library and Beyond" | CppCon 2016 | 2016-09-18 |
C++17’s parallel algorithms library (committee draft) adds execution policies and concrete overloads so you can write std::par_unseq or std::par versions of… | ||
| “Benchmarking C++ Code" | CppCon 2015 | 2015-09-21 |
Explains how to get believable C++ performance numbers: form a hypothesis, measure runtimes with std::chrono or the TSC, and report 95% confidence intervals… | ||
| Practical Functional Programming in C++ | CppCon 2014 | 2014-09-07 |
Argues that functional techniques are practical in modern C++ by showing actual code: uses std::function and C++11 lambdas to store callbacks in a MUD‑style… | ||
| Boost.Asio and Boost.Serialization, Part II | CppCon 2014 | 2014-09-07 |
Prototype message-passing runtime uses Boost.Asio and Boost.Serialization to send polymorphic actions over TCP, implements a zero-copy archive (two‑pass… | ||
| Boost.Asio and Boost.Serialization, Part I | CppCon 2014 | 2014-09-07 |
A concrete prototype message system uses Boost.Asio for networking and Boost.Serialization for object transmission, implementing an echo server, two-thread… | ||
| nothing matches | ||