Optimize Automatic Differentiation Performance in C++ - Steve Bronder - CppCon 2025
https://cppcon.org
---
Optimize Automatic Differentiation Performance in C++ - Steve Bronder - CppCon 2025
---
Reverse‑mode automatic differentiation (AD) powers everything from back‑propagation that trains trillion‑parameter large language models to the Stan programming language's Bayesian inference engines. Performance tricks like arena allocators, expression‑templates, SIMD friendly data structures, and template meta-programming finds its way inside C++ AD libraries. Milliseconds saved per gradient compound can turn into hours of wall‑time wins.
This session dissects the engineering behind those performance wins, showcasing improvements across different C++ AD libraries over time. Attendees will see how contemporary C++ AD techniques can make AD so fast that there is rarely a need to add hand-written derivatives to your program. The talk assumes familiarity with modern C++ but no prior exposure to automatic differentiation.
---
Slides: https://github.com/CppCon/CppCon2025/...
Work at Hudson River Trading (HRT): https://tinyurl.com/safxfctf
---
Steve Bronder
Steve Bronder is a software engineer at the Flatiron Institute in New York City who focuses on high‑performance C++ libraries for statistical modeling and automatic differentiation. He is a a core maintainer of the Stan Math project and contributes to open‑source tooling that accelerates scientific computing. Previously, he worked as a data scientist at Capital One. Steve holds an M.A. in Quantitative Methods in the Social Sciences from Columbia University. In his spare time he enjoys tabletop games and running.
---
CppCon is the annual, week-long face-to-face gathering for the entire C++ community. The conference is organized by the C++ community for the community. You will enjoy inspirational talks and a friendly atmosphere designed to help attendees learn from each other, meet interesting people, and generally have a stimulating experience. Taking place this year in Aurora, Colorado, near the Denver airport, and including multiple diverse tracks, the conference will appeal to anyone from C++ novices to experts.
Annual CppCon Conference - https://www.cppcon.org
/ cppcon
https://x.com/cppcon
/ cppconference
/ cppcon
https://mastodon.social/@CppCon
---
Videos Filmed & Edited by Bash Films: http://www.BashFilms.com
YouTube Channel Managed by Digital Medium Ltd: https://events.digital-medium.co.uk
---
#cpp #cplusplus #cppcon #cppprogramming #cplusplusprogramming #softwaredevelopment #softwareengineering #coding #code #computerscience #technology #technews #programming #programmer
On this page of the site you can watch the video online Optimize Automatic Differentiation Performance in C++ - Steve Bronder - CppCon 2025 with a duration of hours minute second in good quality, which was uploaded by the user CppCon 01 December 2025, share the link with friends and acquaintances, this video has already been watched 5,675 times on youtube and it was liked by 139 viewers. Enjoy your viewing!
