Home/2025
Archive

Year 2025

1,454 articles published in 2025.

Security — Scaling Privacy Infrastructure for GenAI Product Innovation

Scaling Privacy Infrastructure for GenAI Product Innovation

How does Meta empower its product teams to harness GenAI’s power responsibly? In this post, we delve into how Meta addresses the challenges of safeguarding data in the GenAI era by scaling its Privacy Aware Infrastructure (PAI), with a particular focus on Meta’s AI glasses as an example GenAI use case. We’ll describe in detail […]

CWChris Wiltz·October 23, 2025Security
Engineering — Advancing Our Chef Infrastructure: Safety Without Disruption

Advancing Our Chef Infrastructure: Safety Without Disruption

Last year, I wrote a blog post titled Advancing Our Chef Infrastructure, where we explored the evolution of our Chef infrastructure over the years. We talked about the shift from a single Chef stack to a multi-stack model, and the challenges that came with it – from updating how we handle cookbook uploads to navigating…

AGArchie Gunasekara·October 23, 2025Engineering
Performance — A deep dive into BPF LPM trie performance and optimization

A deep dive into BPF LPM trie performance and optimization

This post explores the performance of BPF LPM tries, a critical data structure used for IP matching. It delves into the trie algorithm, presents benchmark data, and explains how the current kernel implementation’s design choices lead to performance bottlenecks.

MFMatt Fleming, Jesper BrouerMatt Fleming, Jesper Brouer·October 21, 2025Performance
Performance — Benefits of a DESCending index

Benefits of a DESCending index

The DESC clause in CREATE INDEX is rarely needed. I'll show use cases for descending indexes, including storage efficiency and performance.

LALaurenz Albe·October 21, 2025Performance
AI & ML — How to update community health files with AI

How to update community health files with AI

Have you ever thought about using AI to update community health files for your repositories? This blog shares actionable next steps for doing just that, including a starter kit with a checklist and tutorials on how to create three useful files.

ALAlexandra LietzkeAlexandra Lietzke·October 21, 2025AI & ML
Engineering — Building a Honeypot Field That Works

Building a Honeypot Field That Works

Honeypots are fields that developers use to prevent spam submissions. They still work in 2025. But you got to set a couple of tricks in place so spambots can’t detect your honeypot field.

ZLZell LiewZell Liew·October 20, 2025Engineering
SRE & Ops — Disaggregated Scheduled Fabric: Scaling Meta’s AI Journey

Disaggregated Scheduled Fabric: Scaling Meta’s AI Journey

Disaggregated Schedule Fabric (DSF) is Meta’s next-generation network fabric technology for AI training networks that addresses the challenges of existing Clos-based networks. We’re sharing the challenges and innovations surrounding DSF and discussing future directions, including the creation of mega clusters through DSF and non-DSF region interconnectivity, as well as the exploration of alternati

CWChris Wiltz·October 20, 2025SRE & Ops
AI & ML — Scaling LLM Inference: Innovations in Tensor Parallelism, Context Parallelism, and Expert Parallelism

Scaling LLM Inference: Innovations in Tensor Parallelism, Context Parallelism, and Expert Parallelism

At Meta, we are constantly pushing the boundaries of LLM inference systems to power applications such as the Meta AI App. We’re sharing how we developed and implemented advanced parallelism techniques to optimize key performance metrics related to resource efficiency, throughput, and latency. The rapid evolution of large language models (LLMs) has ushered in a […]

PHPhil Hornshaw·October 17, 2025AI & ML
AI & ML — AI In UX: Achieve More With Less — Smashing Magazine

AI In UX: Achieve More With Less — Smashing Magazine

A simple but powerful mental model for working with AI: treat it like an enthusiastic intern with no real-world experience. Paul Boag shares lessons learned from real client projects across user research, design, development, and content creation.

PBPaul BoagPaul Boag·October 17, 2025AI & ML
SRE & Ops — Branching in a Sapling Monorepo

Branching in a Sapling Monorepo

Sapling is a scalable, user-friendly, and open-source source control system that powers Meta’s monorepo. As discussed at the GitMerge 2024 conference session on branching, designing and implementing branching workflows for large monorepos is a challenging problem with multiple tradeoffs between scalability and the developer experience. After the conference, we designed, implemented, and open sourc

CWChris Wiltz·October 16, 2025SRE & Ops
SRE & Ops — 10X Backbone: How Meta Is Scaling Backbone Connectivity for AI

10X Backbone: How Meta Is Scaling Backbone Connectivity for AI

We’re sharing details on our journey to scale Meta’s Backbone network to support the increasing demands of new and existing AI workloads. We’ve developed new technologies and designs to address our 10x scaling needs and applying some of these same principles to help extend our AI clusters between multiple data centers. Meta’s Backbone network is […]

PHPhil Hornshaw·October 16, 2025SRE & Ops
Performance — Unpacking Cloudflare Workers CPU Performance Benchmarks

Unpacking Cloudflare Workers CPU Performance Benchmarks

Cloudflare investigated CPU performance benchmark results for Workers, uncovering and fixing issues in infrastructure, V8 garbage collection, and OpenNext optimizations. These improvements have made Cloudflare Workers faster for all customers.

KVKenton VardaKenton Varda·October 14, 2025Performance
SRE & Ops — Design for Sustainability: New Design Principles for Reducing IT Hardware Emissions

Design for Sustainability: New Design Principles for Reducing IT Hardware Emissions

We’re presenting Design for Sustainability, a set of technical design principles for new designs of IT hardware to reduce emissions and cost through reuse, extending useful life, and optimizing design. At Meta, we’ve been able to significantly reduce the carbon footprint of our data centers by integrating several design strategies such as modularity, reuse, retrofitting, […]

CWChris Wiltz·October 14, 2025SRE & Ops
Engineering — Introducing stablecoin payments for subscriptions

Introducing stablecoin payments for subscriptions

As businesses expand globally, stablecoins can help them reach a broader international audience more cost-effectively. Now we’re extending these benefits to subscription-based businesses by launching recurring payments with stablecoins.

JLJennifer LeeJennifer Lee·October 14, 2025Engineering
Frontend — Masonry: Watching a CSS Feature Evolve

Masonry: Watching a CSS Feature Evolve

What can CSS Masonry discussions teach us about the development of new CSS features? What is the CSSWG’s role? What influence do browsers have? What can learn from the way past features evolved?

SMSaleh MubasharSaleh Mubashar·October 13, 2025Frontend
AI & ML — GitHub Copilot CLI: How to get started

GitHub Copilot CLI: How to get started

Discover how to use GitHub Copilot directly in the terminal. From cloning a repository to opening a pull request, here’s how to streamline your workflow with Copilot CLI.

AGAndrea GriffithsAndrea Griffiths·October 13, 2025AI & ML
SRE & Ops — OCP Summit 2025: The Open Future of Networking Hardware for AI

OCP Summit 2025: The Open Future of Networking Hardware for AI

At Open Compute Project Summit (OCP) 2025, we’re sharing details about the direction of next-generation network fabrics for our AI training clusters. We’ve expanded our network hardware portfolio and are contributing new disaggregated network platforms to OCP. We look forward to continued collaboration with OCP to open designs for racks, servers, storage boxes, and motherboards […]

CWChris Wiltz·October 13, 2025SRE & Ops
AI & ML — The Grayscale Problem — Smashing Magazine

The Grayscale Problem — Smashing Magazine

From A/B tests to AI slop, the modern web is bleeding out its colour. Standardized, templated, and overoptimized, it’s starting to feel like a digital Levittown. But it doesn’t have to be.

FOFrederick O BrienFrederick O Brien·October 13, 2025AI & ML
Security — Introducing REACT: Why We Built an Elite Incident Response Team

Introducing REACT: Why We Built an Elite Incident Response Team

We're launching Cloudforce One REACT, a team of expert security responders designed to eliminate the gap between perimeter defense and internal incident response. The team brings unmatched threat intelligence and network-native mitigation to help organizations prepare for and respond to security crises across any environment.

COChris Orourke, Utsav AdhikariChris Orourke, Utsav Adhikari·October 9, 2025Security