![[Uber Open Source ] Distributed Deep Learning with Horovod Alex Sergeev @UberEngineering](https://i.ytimg.com/vi/D1By2hy4Ecw/maxresdefault.jpg)
Uploaded March 2019 | Updated September 2026, 2 weeks ago
During a March 2019 meetup hosted at Uber's SF office, Uber senior software engineer Alex Sergeev discusses how frameworks like TensorFlow, PyTorch, and MXNet make it easy to design and train deep learning models. However, when it comes to scaling models to multiple GPUs in a server, or multiple servers in a cluster, difficulties usually arise.
During a March 2019 meetup hosted at Uber's SF office, Uber senior software engineer Alex Sergeev discusses how frameworks like TensorFlow, PyTorch, and MXNet make it easy to design and train deep learning models. However, when it comes to scaling models to multiple GPUs in a server, or multiple servers in a cluster, difficulties usually arise.
![[Uber Open Source ] Distributed Deep Learning with Horovod Alex Sergeev](https://i.ytimg.com/vi/D1By2hy4Ecw/mqdefault.jpg "[Uber Open Source ] Distributed Deep Learning with Horovod Alex Sergeev
During a March 2019 meetup hosted at Ubers SF office, Uber senior software engineer Alex Sergeev discusses how frameworks like TensorFlow, PyTorch, and MXNet make it easy to design and train deep learning models. However, when it comes to scaling models to multiple GPUs in a server, or multiple servers in a cluster, difficulties usually arise.")
![[Uber Open Summit 2018] Scaling a React Component Library: Building Uber’s new Design Language](https://i.ytimg.com/vi/DmHlOvEGdfk/mqdefault.jpg "[Uber Open Summit 2018] Scaling a React Component Library: Building Uber’s new Design Language
Duplication of code and inconsistent design and experience are major issues across web applications. During this Uber Open Summit 2018 tech talk, engineer Nadiia Dmytrenko discusses the process of designing and building a high-quality library of reusable, composable, themeable, and highly-customizable React component library to create a consistent design.")
![[Destination:Web] Discovering Web Audio by Recreating Deep Note](https://i.ytimg.com/vi/DtqTEZJHeXU/mqdefault.jpg "[Destination:Web] Discovering Web Audio by Recreating Deep Note
During a January 2019 meetup, Facebook software engineer Stoyan Stefanov discusses how to recreate the THXs Deep Note, an audio logo played before most movies start, in JavaScript using the Web Audio API.")
![[Payments Platform] Evolution of Payments at Uber Nimish Sheth & Steven Karis](https://i.ytimg.com/vi/Dz6dAZs8Scg/mqdefault.jpg "[Payments Platform] Evolution of Payments at Uber Nimish Sheth & Steven Karis
Uber’s payments architecture is composed of two main parts: collections and disbursements. In this presentation from a September 2018 meetup, engineers Nimish Sheth & Steven Karis offers a closer look at our high-level payments stack, core data models, and cash money movements.")
![[Uber Open Source] Leveraging Pinot at Uber for Large-scale Analytics](https://i.ytimg.com/vi/E8Xsmc9AqDM/mqdefault.jpg "[Uber Open Source] Leveraging Pinot at Uber for Large-scale Analytics
Apache Pinot is a distributed columnar database for real-time analytics. It powers many well-known applications, including LinkedIn’s Who Viewed My Profile feature and Uber Eats timely analytics. Since Pinot joined Apache Incubator in 2019, it has gained a lot of momentum with a growing community.
During this May 2020 Uber Engineering Meetup, software engineers Haibo Wang and James Shao discuss how Pinot empowers timely analytics at Uber. We will also dive into how we built an open source Presto connector to enable complex SQL queries on Pinot and other data sources, as well as how we developed upserts for Pinot to allow mutable data in real-time tables.
Learn more: https://eng.uber.com/engineering-sql-support-on-apache-pinot/")
![[Uber Eng Blog] Visualizing Uber Air Simulated Flights](https://i.ytimg.com/vi/Edr0Adlka-Y/mqdefault.jpg "[Uber Eng Blog] Visualizing Uber Air Simulated Flights
In this visualization from the Tuesday, October 29, 2019 Uber Engineering Blog article, Taking City Visualization into the Third Dimension with Point Clouds, 3D Tiles, and deck.gl, Uber Air flight routes are simulated using our visualization suite.
Learn more about these powerful data visualizations tools on the Uber Engineering Blog: eng.uber.com")
![[Go NYC] The Myth around Go Frameworks & Libraries Venkatesh Thallam](https://i.ytimg.com/vi/EiBHiMexzEI/mqdefault.jpg "[Go NYC] The Myth around Go Frameworks & Libraries Venkatesh Thallam
During an October 2018 Go Language NYC meetup, Paperless Post engineer Venkatesh Thallam discusses apprehensions using Go frameworks and external libraries and offers suggestions to increase Go development flexibility.")
![[Destination:Web] Fusion.js: A Plugin-based Universal Web Framework](https://i.ytimg.com/vi/Enzc0D8EI_8/mqdefault.jpg "[Destination:Web] Fusion.js: A Plugin-based Universal Web Framework
During this September 2018 Destination:Web meetup, Uber engineer Leo Horie discusses Fusion.js, Ubers open source, plugin-based universal web framework to build lightweight, high-performing web apps.")
![[Uber Open Summit Sofia 2019] Metrics at Uber](https://i.ytimg.com/vi/GYmDVsNueY4/mqdefault.jpg "[Uber Open Summit Sofia 2019] Metrics at Uber
During this Uber Open Summit Sofia 2019 presentation, engineer Celina Ward discusses why we built M3, our open source metrics stack, and M3DB, our open source distributed time series database. She also walks through how to use M3 and the future of its development roadmap.
Learn more about Uber Open Summit Sofia 2019: https://uberopensofia.splashthat.com/")
