InfoQ Homepage Facebook Content on InfoQ
-
0 → 1, Shipping Threads in 5 Months
Zahan Malkani shares how they built a microblogging service to compete with Twitter with a small team that shipped a new social network in a few months.
/presentations/threads-meta/en/mediumimage/ZahanMalkani-medium-1723461991582.jpg)
-
Global Capacity Management through Strategic Demand Allocation
Ranjith Kumar discusses abstractions and guarantees, the design and implementation for managing workloads across 10s of regions, categorizing & modeling, and achieving global capacity management.
/presentations/global-capacity-management/en/mediumimage/RanjithKumarS-medium-1713347613858.jpg)
-
Architecting a Production Development Environment for Reliability
At Meta, developers use servers (devservers) to perform their daily work. This talk discusses their software architecture and the mechanisms employed to ensure they remain reliable and available.
/presentations/meta-production-environment/en/mediumimage/HenriqueAndrade-medium-1702375968305.jpg)
-
How Facebook Is Bringing QUIC to Billions
Matt Joras and Yang Chi discuss the technical challenges implementing QUIC and HTTP/3, from edge load balancer to mobile clients, and from application tweaking to transport congestion control.
/presentations/facebook-quic-http3/en/mediumimage/facebook-quic-http3-m-1642498804884.jpg)
-
Differentiable Programming in Kotlin
Irene Dea discusses how Facebook is using Kotlin, developing a new differentiable programming framework for it.
/presentations/differentiable-framework-kotlin/en/mediumimage/Irene Dea-m-1640191803451.jpg)
-
Probabilistic Programming for Software Engineers
Michael Tingley provides a preview of how Facebook is advancing probabilistic programming, as well as some of the big problems they used it to solve.
/presentations/facebook-ppl/en/mediumimage/Micbig-1589359842416.JPG)
-
From Research to Production with PyTorch
Jeff Smith covers some of the latest features from PyTorch including the TorchScript JIT compiler, distributed data parallel training, TensorBoard integration, new APIs, and more.
/presentations/pytorch-torchscript-botorch/en/mediumimage/Jeff-Smith-m-1566323707652.jpg)
-
wav2letter++: Facebook's Fast Open-Source Speech Recognition System
Vitaliy Liptchinsky introduces wav2letter++, an open-source deep learning speech recognition framework, explaining its architecture and design, and comparing it to other speech recognition systems.
/presentations/wav2letter-facebook/en/mediumimage/Vitbig-1560468358688.JPG)
-
Canopy: Scalable Distributed Tracing & Analysis @ Facebook
Haozhe Gao and Joe O’Neill present Canopy, Facebook’s performance and efficiency tracing infrastructure. They talk about the lessons learned and present case studies of its use.
/presentations/canopy-scalable-tracing-analytics-facebook/en/mediumimage/Doibig-1534298649948.JPG)
-
Applied Performance Theory
Kavya Joshi explores the use of performance theory in real systems at companies like Facebook, and discusses how it can be leveraged, to prepare systems for flux and scale.
/presentations/little-usl-scalability-performance/en/mediumimage/Kavbig-1530319072200.JPG)
-
Scalable Chatbot Architecture with eBay ShopBot
Robert Enyedi discusses the ebay ShopBot, a personal shopping assistant available as a Facebook Messenger bot.
/presentations/ebay-shopbot/en/mediumimage/Robbig.JPG)
-
Machine Learning at Scale
Aditya Kalro discusses using large-scale data for Machine Learning (ML) research and some of the tools Facebook uses to manage the entire process of training, testing, and deploying ML models.
/presentations/machine-learning-scale/en/mediumimage/Adbig.JPG)