Training and serving frontier AI models depends on fast, reliable networks that move data between GPUs without wasting compute cycles.  To meet this challenge at scale, Meta designed MetaRoCE – a clean-sheet RDMA transport protocol purpose-built for AI workloads on commodity Ethernet.  We’re releasi...

MTIA 300 is the first of Meta’s family of in-house training and inference accelerators optimized for training ranking and recommendation models. We’re sharing how MTIA 300’s built-in NIC chiplets allow it to meet the communication needs associated with training recommendation models with superior pe...

WhatsApp is committed to helping people stay safe while protecting the privacy of their messages. As scam tactics evolve — from impersonation to social engineering to AI-generated lures — we’re always evolving as well, so that our protections stay ahead of scammers while protecting people’s personal...

Every day, Meta’s recommendation platforms handle billions of user interactions, generating rich temporal signals that capture individual preferences and intent across products, ads, and content. In our 2024 post on sequence learning for ads recommendations, we showed how modeling the order and timi...

Meta’s Generative Ads Recommendation Model (GEM), the foundation model behind ads recommendations across Instagram and Facebook, now trains at LLM scale on several thousand of the latest-generation GPUs. This post goes into the details on how we achieved: doubling end-to-end (E2E) training efficienc...

Hierarchical Interest Representation is a research area for Meta Ads. We’re exploring an upstream representation layer over the universe of Ads entities – users, advertisers, products, services – learning unified embeddings that connect users’ inferred interests with the breadth of what advertisers ...

TL; DR At Meta’s scale, a few milliseconds of latency degradation can have a significant negative impact on ads performance.  When a Linux kernel upgrade risked regressing latency across Meta’s ad serving fleet, we turned to sched_ext — the upstream, BPF-based extensible scheduling framework — to bu...

Over the past several years, model capabilities and training dataset sizes have experienced exponential growth. During the past year or so, the time between new-frontier-model releases has gone down from months to weeks. Reliable and fast access to storage is important to both the speed and computat...

This year marks Meta’s 10th consecutive year as a sponsor of the Python Software Foundation (PSF), the charitable organization dedicated to advancing, supporting, and protecting the open-source Python programming language and the community that sustains it. Python is one of the world’s most influent...

Privacy controls — systems that enforce retention, access, allowed-purpose, downstream-sharing, or anonymization policies — require a reliable understanding of data to function. Before such a control can operate effectively, it must know exactly what it is looking at. This can be complex, as demonst...

Smart glasses like the Ray-Ban Meta and Oakley Meta Vanguards need to pack enough energy to power features like cameras, speakers, AI workloads, and even a display. But it all has to fit into the glasses’ temple arms. So how do you place a battery with enough power to run a pair of smart glasses [.....

Adopting AV1 for real-time communication at Meta has been a multi-year effort spanning codec selection, device eligibility, rate control, and error resilience. We’re sharing the technical and operational challenges while deploying AV1 and expanding coverage, and how we addressed them for real-time c...

We’re introducing Instantaneous PowerLoss Storm, a new testing paradigm within Meta’s infrastructure for handling and mitigating instant or zero-notice power loss in our data centers.  We’re sharing: how we built readiness to tolerate instant failures into our existing systems with defense-in-depth ...

We’re introducing SilverTorch, a reimagining of recommendation systems that unifies all retrieval components for user generated content under a unified architecture.  SilverTorch shows up to 23.7x higher throughput compared to the state-of-the-art approaches. It’s also showing 20.9x more compute cos...

On its face the new Friend Bubbles feature looks simple enough. It highlights Reels your friends have watched and reacted to. But sometimes the features that seem the most straightforward require the deepest engineering work. On this episode of the Meta Tech Podcast, Pascal Hartig chats with Subasre...

Meta’s data ingestion system, which our engineering teams leverage for up-to-date snapshots of the social graph, has recently undergone a significant revamp to enhance its reliability at scale.  Moving from our legacy system to our new architecture required a large-scale migration of our entire data...

We’re rolling out version 1.1 of Labyrinth, the encrypted storage system and protocol that secures messages and history on Messenger. Labyrinth 1.1 enhances the reliability of end-to-end encrypted backups with a new sub-protocol that helps messages survive the loss of a device, a switched device, an...

The HSM-based Backup Key Vault Meta’s HSM-based Backup Key Vault provides the foundation for end-to-end encrypted backups for WhatsApp and Messenger. The system allows people to protect their backed-up message history with a recovery code, ensuring that the recovery code is stored in tamper-resistan...

We’ve fundamentally transformed Facebook Groups Search to help people more reliably discover, sort through, and validate community content that’s most relevant to them. We’ve adopted a new hybrid retrieval architecture and implemented automated model-based evaluation to address the major friction po...

We’re sharing insights into Meta’s Capacity Efficiency Program, where we’ve built an AI agent platform that helps automate finding and fixing performance issues throughout our infrastructure. By leveraging encoded domain expertise across a unified, standardized tool interface these agents help save ...

We’re sharing lessons learned from Meta’s post-quantum cryptography (PQC) migration to help other organizations strengthen their resilience as industry transitions to post-quantum cryptography standards. We’re proposing the idea of PQC Migration Levels to help teams within organizations manage the c...

At Meta, WebRTC powers real-time audio and video across various platforms. But forking a large open-source project like WebRTC within our monorepo presents unique challenges – over time, an internal fork can drift behind upstream, cutting itself off from community upgrades. We’re sharing how we esca...

As AI increases developer speed and productivity it also increases the need for safeguards. On this episode of the Meta Tech Podcast, Pascal Hartig sits down with Ishwari and Joe from Meta’s Configurations team to discuss how Meta makes config rollouts safe at scale. Listen in to learn about canaryi...

AI coding assistants are powerful but only as good as their understanding of your codebase. When we pointed AI agents at one of Meta’s large-scale data processing pipelines – spanning four repositories, three languages, and over 4,100 files – we quickly found that they weren’t making useful edits qu...

This is the second post in the Ranking Engineer Agent blog series exploring the autonomous AI capabilities accelerating Meta’s Ads Ranking innovation. The previous post introduced Ranking Engineer Agent’s ML exploration capability, which autonomously designs, executes, and analyzes ranking model exp...

Meta continues to lead the industry in utilizing groundbreaking AI Recommendation Systems (RecSys) to deliver better experiences for people, and better results for advertisers. To reach the next frontier of performance, we are scaling Meta’s Ads Recommender runtime models to LLM-scale & complexity t...

Meta is continuing its long-term roadmap to help the construction industry leverage AI to produce high-quality and more sustainable concrete mixes, as well as those exclusively produced in the United States.  Concurrent with the 2026 American Concrete Institute (ACI) Spring Convention, Meta is relea...

Friend bubbles in Facebook Reels highlight Reels your friends have liked or reacted to, helping you discover new content and making it easier to connect over shared interests. This article explains the technical architecture behind friend bubbles, including how machine learning estimates relationshi...

Meta’s Ranking Engineer Agent (REA) autonomously executes key steps across the end-to-end machine learning (ML) lifecycle for ads ranking models. This post covers REA’s ML experimentation capabilities: autonomously generating hypotheses, launching training jobs, debugging failures, and iterating on ...

Even seemingly simple engineering tasks — like updating an API — can become monumental undertakings when you’re dealing with millions of lines of code and thousands of engineers, especially if the changes are security-related. Nowhere is this more apparent than in mobile security, where a single cla...