All blogs / Distributed systems / cloud
Engineering at Meta
Engineering at Meta Blog
Distributed systems / cloudAI / MLNetworkingEngineering blogs
- Posts
- 1,142 from 14 May 2008 to 29 Oct 2026
- Posting
- weekly (74 in the last year)
- Kind
- company blog
- Feed
- https://engineering.fb.com/feed/
2026 61
- Catching Bugs Before They Land: JiTTesting’s Human-on-the-Loop Approach at Full Scale
- Meta Andromeda: Powering the Next Generation of Ads Retrieval Through Model-System Co-Design
- PrivacyLib: Privacy Controls at the Data Access Boundary
- Python Typing Survey 2026: AI Tools and the Shifting Checker Landscape
- Meta Networking at the 2026 OCP Global Summit: Open Innovation for Scale-Up, Scale-Out, and Scale-Across AI Networks
- GenRecs Tokenization: Transforming Ads Recommendation at Meta with Stable Model Vocabulary
- The Thin Orchestrator: How We Made Our AI Agents More Reliable by Giving the Router Less to Do
- Inside Prometheus: Meta’s Journey to a Gigawatt-Scale Supercluster
- How Meta Built Forum With AI-Native Engineering
- 2026 World Cup: Breaking Records, Spiky Charts and Reliability at Scale
- Sapienz Receives ISSTA 2026 Impact Paper Award
- Meta’s Generative Ads Model (GEM): Scaling the Central Model and Compute Engine for Frontier Ads AI
- OCP Summit 2026: Powering AI with Open Innovations in Scale-Up, Scale-Out, and Scale-Across Networks
- Shipping Media Over QUIC at Meta Scale for Instagram Live Ingest
- Turning Opportunistic AI Inference Capacity Into Better Ads Recommendations
- NTS: Authenticated Time at Meta
- Bringing Private Processing to Meta AI Glasses
- Open-Sourcing Rebalancer: A Generic, High-Performance Library for Solving Assignment Problems
- Inside Petal: Building the World’s First Petabit-Class Transoceanic Subsea Cable
- ZGateway: Learnings from Putting a Proxy in Front of ZippyDB
- An Organizational Second Brain: Building an AI That Learns From Experts
- MetaRoCE: A New RDMA Transport Built for AI-Scale Ethernet
- MTIA 300: Meta’s First Training Chip with Built-in NICs and Communication-Offloading Engines
- How We’re Building Scam Alert on WhatsApp With End-to-End Encryption and Verifiability Guarantees
- From User Sequences to Scaling Laws: A Multi-Stage Architecture for Meta’s Ads Ranking
- GEM Training: How Meta Doubled the Efficiency of Its LLM-Scale Ads Foundation Model
- Exploring Hierarchical Interest Representation For Meta Ads Deep Funnel Optimization
- Modernizing the Meta Ads Service With an Open-Source Kernel Scheduler
- Meta’s AI Storage Blueprint at Scale
- 10 Years of Meta’s Commitment to Python
- Privacy-Aware Infrastructure in the AI-Native Era: An Asset Classification Case Study
- How Meta Engineered Ultra-Narrow Batteries for AI Glasses
- Adopting AV1 for Real-Time Communication (RTC) at Scale
- Lights Out, Systems On: Validating Instant Power Loss Readiness
- SilverTorch: Index as Model — A New Retrieval Paradigm for Recommendation Systems
- Reel Friends: Building Social Discovery that Scales to Billions
- Migrating Data Ingestion Systems at Meta Scale
- Labyrinth 1.1: Making End-to-End Encrypted Backups Even More Reliable
- How Meta Is Strengthening End-to-End Encrypted Backups
- Modernizing the Facebook Groups Search to Unlock the Power of Community Knowledge
- Capacity Efficiency at Meta: How Unified AI Agents Optimize Performance at Hyperscale
- Post-Quantum Cryptography Migration at Meta: Framework, Lessons, and Takeaways
- Escaping the Fork: How Meta Modernized WebRTC Across 50+ Use Cases
- Trust But Canary: Configuration Safety at Scale
- How Meta Used AI to Map Tribal Knowledge in Large-Scale Data Pipelines
- KernelEvolve: How Meta’s Ranking Engineer Agent Optimizes AI Infrastructure
- Meta Adaptive Ranking Model: Bending the Inference Scaling Curve to Serve LLM-Scale Models for Ads
- AI for American-Produced Cement and Concrete
- Friend Bubbles: Enhancing Social Discovery on Facebook Reels
- Ranking Engineer Agent (REA): The Autonomous AI Agent Accelerating Meta’s Ads Ranking Innovation
- Patch Me If You Can: AI Codemods for Secure-by-Default Android Apps
- How Advanced Browsing Protection Works in Messenger
- FFmpeg at Meta: Media Processing at Scale
- Investing in Infrastructure: Meta’s Renewed Commitment to jemalloc
- RCCLX: Innovating GPU Communications on AMD Platforms
- The Death of Traditional Testing: Agentic Development Broke a 50-Year-Old Field, JiTTesting Can Revive It
- Building Prometheus: How Backend Aggregation Enables Gigawatt-Scale AI Clusters
- No Display? No Problem: Cross-Device Passkey Authentication for XR Devices
- Rust at Scale: An Added Layer of Security for WhatsApp
- Adapting the Facebook Reels RecSys AI Model Based on User Feedback
- CSS at Scale With StyleX
2025 70
- Python Typing Survey 2025: Code Quality and Flexibility As Top Reasons for Typing Adoption
- DrP: Meta’s Root Cause Analysis Platform at Scale
- How We Built Meta Ray-Ban Display: From Zero to Polish
- How AI Is Transforming the Adoption of Secure-by-Default Mobile Frameworks
- Zoomer: Powering AI Performance at Meta’s Scale Through Intelligent Debugging and Optimization
- Key Transparency Comes to Messenger
- Efficient Optimization With Ax, an Open Platform for Adaptive Experimentation
- Announcing the Completion of the Core 2Africa System: Building the Future of Connectivity Together
- Enhancing HDR on Instagram for iOS With Dolby Vision
- Open Source Is Good for the Environment
- StyleX: A Styling Library for CSS at Scale
- Meta’s Generative Ads Model (GEM): The Central Brain Accelerating Ads Recommendation AI Innovation
- Video Invisible Watermarking at Scale
- Scaling Privacy Infrastructure for GenAI Product Innovation
- Disaggregated Scheduled Fabric: Scaling Meta’s AI Journey
- Scaling LLM Inference: Innovations in Tensor Parallelism, Context Parallelism, and Expert Parallelism
- Branching in a Sapling Monorepo
- 10X Backbone: How Meta Is Scaling Backbone Connectivity for AI
- Design for Sustainability: New Design Principles for Reducing IT Hardware Emissions
- How Meta Is Leveraging AI To Improve the Quality of Scope 3 Emission Estimates for IT Hardware
- OCP Summit 2025: The Open Future of Networking Hardware for AI
- Introducing the React Foundation: The New Home for React & React Native
- Introducing OpenZL: An Open Source Format-Aware Compression Framework
- Introducing the Candle Subsea Cable, Updates to Our Asia-Pacific Connectivity Projects
- Accelerating our Android apps with Baseline Profiles
- LLMs Are the Key to Mutation Testing and Better Compliance
- Meta 3D AssetGen: Generating 3D Worlds With AI
- Meta’s Infrastructure Evolution and the Advent of AI
- Networking at the Heart of AI — @Scale: Networking 2025 Recap
- Video Streaming With the AV1 Video Codec in Mobile Devices
- Read Meta’s 2025 Sustainability Report
- A New Ranking Framework for Better Notification Quality on Instagram
- Enabling Kotlin incremental compilation on Buck2
- Creating AI agent solutions for warehouse data access and security
- Federation Platform and Privacy Waves: How Meta distributes compliance-related tasks at scale
- Diff Risk Score: AI-driven risk-aware software development
- Building a human-computer interface for everyone
- Accelerating on-device ML on Meta’s family of apps with ExecuTorch
- Policy Zones: How Meta enforces purpose limitation at scale in batch processing systems
- How Meta keeps its AI hardware reliable
- Using AI to make lower-carbon, faster-curing concrete
- An inside look at Meta’s transition from C to Rust on mobile
- Meta joins Kotlin Foundation
- Extending the Malbec subsea cable to Southern Brazil
- Journey to 1000 models: Scaling Instagram’s recommendation system
- Meta’s Full-stack HHVM optimizations for GenAI
- Open-sourcing Pyrefly: A faster Python type checker written in Rust
- Introducing Pyrefly: A new type checker and IDE experience for Python
- Accelerating GPU indexes in Faiss with NVIDIA cuVS
- Enhancing the Python ecosystem with type checking and free threading
- Taking the plunge: The engineering journey of building a subsea cable
- Introducing AutoPatchBench: A Benchmark for AI-Powered Security Fixes
- Building Private Processing for AI tools on WhatsApp
- How Meta understands data at scale
- Meta Open Source: 2024 by the numbers
- Mobile GraphQL at Meta in 2025
- Building multimodal AI for Ray-Ban Meta glasses
- A case for QLC SSDs in the data center
- How Meta is translating its Java codebase to Kotlin
- Protecting user data through source code analysis at scale
- Unlocking global AI potential with next-generation subsea infrastructure
- Looking back at our Bug Bounty program in 2024
- Revolutionizing software testing: Introducing LLM-powered bug catchers
- Data logs: The latest evolution in Meta’s access tools
- How Precision Time Protocol handles leap seconds
- Bringing Jetpack Compose to Instagram for Android
- How Meta discovers data flows via lineage at scale
- Strobelight: A profiling service built on open source technology
- Measuring productivity impact with Diff Authoring Time
- ILA Evo: Meta’s journey to reimagine fiber optic in-line amplifier sites
2024 62
- Indexing code at scale with Glean
- Translating Java to Kotlin at Scale
- How we think about Threads’ iOS performance
- How to build a mixed reality headset
- Inside Facebook’s video delivery system
- Typed Python in 2024: Well adopted, yet usability challenges persist
- Powering AI innovation by acccelerating the next wave of nuclear
- Meta Andromeda: Supercharging Advantage+ automation with the next-gen personalized ads retrieval engine
- Sequence learning: A paradigm shift for personalized ads recommendations
- How Meta built large-scale cryptographic monitoring
- Diff Authoring Time: Measuring developer productivity at Meta
- IPLS: Privacy-preserving storage for your WhatsApp contacts
- OCP Summit 2024: The open future of networking hardware for AI
- Meta’s open AI hardware vision
- How open source AI can improve population estimates, sustainable energy, and the delivery of climate change interventions
- React at Meta Connect 2024
- Inside Bento: Jupyter Notebooks at Meta
- Simulator-based reinforcement learning for data center cooling optimization
- Read Meta’s 2024 Sustainability Report
- Meta is getting ready for post-quantum cryptography
- How Meta enforces purpose limitation via Privacy Aware Infrastructure at scale
- RETINAS: Real-Time Infrastructure Accounting for Sustainability
- How PyTorch powers AI training and inference
- Inside the hardware and co-design of MTIA
- Bringing Llama 3 to life
- Aparna Ramani discusses the future of AI infrastructure
- How Meta animates AI-generated images at scale
- RoCE networks for distributed AI training at scale
- DCPerf: An open source benchmark suite for hyperscale compute applications
- Meet Caddy – Meta’s next-gen mixed reality CAD software
- AI Lab: The secrets to keeping machine learning engineers moving fast
- Taming the tail utilization of ads inference at Meta scale
- Meta’s approach to machine learning prediction robustness
- The key to a happy Rust/C++ relationship
- Leveraging AI for efficient incident response
- PVF: A novel metric for understanding AI systems’ vulnerability against SDCs in model parameters
- MLow: Meta’s low bitrate audio codec
- How Meta trains large language models at scale
- Maintaining large-scale AI capacity at Meta
- Unlocking the power of mixed reality devices with MobileConfig
- Serverless Jupyter Notebooks at Meta
- Composable data management at Meta
- Post-quantum readiness for TLS at Meta
- Behind the scenes of Threads for web
- Building new custom silicon for Meta’s AI workloads
- Building an infrastructure for AI’s future
- Introducing the next-gen Meta Training and Inference Accelerator
- Bringing HDR photo support to Instagram and Threads
- Threads has entered the fediverse
- Optimizing RTC bandwidth estimation with machine learning
- Better video for mobile RTC with AV1 and HD
- Logarithm: A logging engine for AI training workflows and services
- Building Meta’s GenAI Infrastructure
- Making messaging interoperability with third parties safe for users in Europe
- How DotSlash makes executable deployment simpler
- Aligning Velox and Apache Arrow: Towards composable data management
- Meta loves Python
- Simple Precision Time Protocol at Meta
- DotSlash: Simplified executable deployment
- Improving machine learning iteration speed with faster application build and packaging
- Lazy is the new fast: How Lazy Imports and Cinder accelerate machine learning at Meta
- How Meta is advancing GenAI
2023 49
- How Meta built the infrastructure for Threads
- AI debugging at Meta with HawkEye
- Building end-to-end security for Messenger
- Writing and linting Python at scale
- Watch: Meta’s engineers on building network infrastructure for AI
- Enhancing the security of WhatsApp calls
- How Meta built Threads in 5 months
- Automating data removal
- Automating dead code cleanup
- 5 Things you didn’t know about Buck2
- How Meta is creating custom silicon for AI
- Automating product deprecation
- Meta contributes new features to Python 3.12
- Meta Quest 2: Defense through offense
- Using Chakra execution traces for benchmarking and network performance optimization
- Arcadia: An end-to-end AI system performance simulator
- Threads: The inside story of Meta’s newest social app
- What is it like to write code at Meta?
- Scheduling Jupyter Notebooks at Meta
- Code Llama: Meta’s state-of-the-art LLM for coding
- Introducing Immortal Objects for Python
- Meta Connect 2023: September 27 – 28
- Scaling the Instagram Explore recommendations system
- How Meta is improving password security and preserving privacy
- Fixit 2: Meta’s next-generation auto-fixing linter
- Using short-lived certificates to protect TLS secrets
- Bringing HDR video to Reels
- Meta’s Evenstar is transitioning to OCP to accelerate open RAN adoption
- Meta developer tools: Working at scale
- Bombyx is being licensed for product development
- MSVP is Meta’s first video processing ASIC
- Meta introduces its first-generation AI inference accelerator
- Building and deploying MySQL Raft at Meta
- The malware threat landscape: NodeStealer, DuckTail, and more
- A fine-grained network traffic analysis with Millisampler
- Deploying key transparency at WhatsApp
- How Device Verification protects your WhatsApp account
- Why xHE-AAC is being embraced at Meta
- Build faster with Buck2: Our open source build system
- Introducing Velox: An open source unified execution engine
- Meta’s head of AR hardware on the future of AR
- How Meta brought AV1 to Reels
- Inside Meta’s first smart glasses
- Building a cross-platform runtime for AR
- Improving Meta’s global maps
- The evolution of Facebook’s iOS app architecture
- Asynchronous computing at Meta: Overview and learnings
- Watch Meta’s engineers discuss optimizing large-scale networks
- Tulip: Modernizing Meta’s data platform
2022 49
- Open-sourcing Anonymous Credential Service
- Enabling static analysis of SQL queries at Meta
- How to create impact through Meta’s research intern program
- Retrofitting null-safety onto Java at Meta
- PTP: Timing accuracy and precision for the future of computing
- How Precision Time Protocol is being deployed at Meta
- Move faster, wait less: Improving code review time at Meta
- Sapling: Source control that’s user-friendly and scalable
- Tulip: Schematizing Meta’s data platform
- Reducing Instagram’s basic video compute time by 94 percent
- Improving Instagram notification management with machine learning and causal inference
- From zero to 10 million lines of Kotlin
- OCP Summit 2022: Open hardware for AI infrastructure
- Launching a new Chromium-based WebView for Android
- Scaling data ingestion for machine learning training at Meta
- How thermal simulation helps optimize Meta’s data centers
- MemLab: An open source framework for finding JavaScript memory leaks
- Open-sourcing TAOBench: An end-to-end social network benchmark
- Network Entitlement: A contract-based network sharing solution
- Viewing the world as a computer: Global capacity management
- Improving Meta’s SLO workflows with data annotations
- Five security principles for billions of messages across Meta’s apps
- Programming languages endorsed for server-side use at Meta
- Launching Instagram Messaging on desktop
- It’s time to leave the leap second in the past
- Using Hermes’s Quicksort to run Doom: A tale of JavaScript exploitation
- How Meta and the security industry collaborate to secure the internet
- Building text animations for Instagram Stories
- Owl: Distributing content at Meta scale
- Watch Meta’s engineers discuss QUIC and TCP innovations for our network
- Transparent memory offloading: more memory at a fraction of the cost and power
- Applying federated learning to protect data on mobile devices
- Under the hood: Meta’s cloud gaming infrastructure
- Introducing Zelos: A ZooKeeper API leveraging Delos
- Cache made consistent
- Meta Open Source is transferring Jest to the OpenJS Foundation
- Language packs: Meta’s mobile localization solution
- BellJar: A new framework for testing system recoverability at scale
- Delta: A highly available, strongly consistent storage service using chain replication
- How the Cinder JIT’s function inliner helps us optimize Instagram
- A new way to survey potential fiber routes — without access to paved roads
- SQL Notebooks: Combining the power of Jupyter and SQL editors for data analytics
- How Meta enables de-identified authentication at scale
- Detecting silent errors in the wild: Combining two novel approaches to quickly detect silent data corruptions at scale
- VESPA: Static profiling for binary optimization
- Code Verify: An open source browser extension for verifying code authenticity on the web
- An open source compositional deadlock detector for Android Java
- Augmenting Flexible Paxos in LogDevice to improve read availability
- FOQS: Making a distributed priority queue disaster-ready
2021 51
- Power Loss Siren: Making Meta resilient to power loss events
- Charting the future of our bug bounty program
- SLICK: Adopting SLOs for improved reliability
- Malbec subsea cable connects Argentina and Brazil with the rest of the world
- OCP Summit 2021: Open networking hardware lays the groundwork for the metaverse
- Kangaroo: A new flash cache optimized for tiny objects
- Autonomous testing of services at scale
- Facebook engineers receive 2021 IEEE Computer Society Cybersecurity Award for static analysis tools
- RTMP Go Away: Lossless reconnections for live streaming
- Inside the Lab: Expanding connectivity by sea, land, and air
- More details about the October 4 outage
- Update about the October 4th outage
- Open-sourcing Mariana Trench: Analyzing Android and Java app security in depth
- 2Africa Pearls subsea cable connects Africa, Europe, and Asia to bring affordable, high-speed internet to 3 billion people
- Superpack: Pushing the limits of compression in Facebook’s mobile apps
- How WhatsApp is enabling end-to-end encrypted backups
- CacheLib, Facebook’s open source caching engine for web-scale services
- RAMP-TAO: Layering atomic transactions on Facebook’s online graph store
- Apricot subsea cable will boost internet capacity, speeds in the Asia-Pacific region
- Open-sourcing a more precise time appliance
- Risk-driven backbone management during COVID-19 and beyond
- How we built a general purpose key value store for Facebook with ZippyDB
- Open sourcing Winterfell: A STARK prover and verifier
- A linear programming approach for optimizing features in ML models
- Migrating Facebook to MySQL 8.0
- Fully Sharded Data Parallel: faster AI training with fewer GPUs
- How WhatsApp enables multi-device capability
- Enforcing encryption at scale
- Ribbon filter: Practically smaller than Bloom and Xor
- Asicmon: A platform agnostic observability system for AI accelerators
- Driving towards an open internet ecosystem to help tackle the digital divide
- Consolidating Facebook storage infrastructure with Tectonic file system
- Meet Kats — a one-stop shop for time series analysis
- Network hose: Managing uncertain network demand with model simplicity
- How Facebook deals with PCIe faults to keep our data centers running reliably
- Peering automation at Facebook
- Running Border Gateway Protocol in large-scale data centers
- Most of computing’s carbon emissions are coming from manufacturing and infrastructure
- A brief history of Rust at Facebook
- Reverse debugging at scale
- DIT — enabling de-identified data collection on WhatsApp
- How Facebook encodes your videos
- Advancing connectivity between the Asia-Pacific region and North America
- Cutting Threads’ send latency in half
- Boosting the performance of virtual machines with Jump-Start
- Mitigating the effects of silent data corruption at scale
- FOQS: Scaling a distributed priority queue
- Faster, more efficient systems for finding and fixing regressions
- Minesweeper automates root cause analysis as a first-line defense against bugs
- Open-sourcing Thrift for Haskell
- How machine learning powers Facebook’s News Feed ranking algorithm
2020 47
- 2020 year in review: Connectivity innovations, faster apps, and progress toward net zero
- A smaller, faster video calling library for our apps
- Evenstar: The road to an alternative infra supply chain for RAN
- On the value of diversified recommendations
- Co-experiences: Hanging out in the digital living room
- Infer powering Microsoft’s Infer#, a new static analyzer for C#
- How Instagram suggests new content
- How do you test your tests?
- How Facebook keeps its large-scale infrastructure hardware up and running
- SuperCell: Reaching new heights for wider connectivity
- The economic impact of subsea cables in Africa
- FioSynth: A representative I/O benchmark and data visualizer for data center workloads
- Building a ubiquitous shared infrastructure using Twine
- How we scale Live streaming for millions of viewers simultaneously
- PPL Bench: Creating a standard for benchmarking probabilistic programming languages
- How Facebook is bringing QUIC to billions
- Introducing resctl-demo: Better resource control with simulation
- Nemo: Data discovery at Facebook
- CG/SQL: Easy, accurate code generation for SQLite
- Mark Harman elected Fellow of the Royal Academy of Engineering
- The next decade: How Facebook is stepping up the fight against climate change
- Throughput autoscaling: Dynamic sizing for Facebook.com
- Fault tolerance through optimal workload placement
- Results of Oregon project environmental analysis
- Containerizing ZooKeeper with Twine: Powering container orchestration from within
- Scaling services with Shard Manager
- Asynchronous computing @Facebook: Driving efficiency and developer productivity at Facebook scale
- DELF: Safeguarding deletion correctness in online social networks
- Pysa: An open source static analysis tool to detect and prevent security issues in Python code
- Pcicrawler: A Python-based command-line interface tool to debug PCI issues at scale
- Making Facebook.com accessible to as many people as possible
- Scalable data classification for security and privacy
- Making aerial fiber deployment faster and more efficient
- Private matching for compute: New solutions to the problem of enabling compute on private set intersections
- Introducing our 2019 Sustainability Report
- Retrie: Haskell refactoring made easy
- Building a transformative subsea cable to better connect Africa
- Accelerometer and SoftSKU: Improving hardware platform performance for diverse microservices
- Rebuilding our tech stack for the new Facebook.com
- Implementing a secure web-based proxy service for Discover
- Turbine: Facebook’s service management platform for stream processing
- Building a more accurate time service at Facebook scale
- Preventing performance regressions with Health Compass and Incident Tracker
- Project LightSpeed: Rewriting the Messenger codebase for a faster, smaller, and simpler messaging app
- Accelerating innovations in infrastructure and advancing global connectivity with our partners
- ZooKeeper Meetup@Facebook: Advancing the state of distributed coordination
- Open source year in review
2019 89
- Systems @Scale Tel Aviv 2019 recap
- Fighting Abuse @Scale 2019 recap
- Data Transfer Project: Enabling portability of photos and videos between services
- CCSM: Scalable statistical anomaly detection to resolve app crashes faster
- Networking @Scale Boston 2019 recap
- Groundbreaking event for our newest long-haul fiber route
- Evaluating COPA congestion control for improved video performance
- TIP Summit 2019: Collaborating to advance internet infrastructure
- Fast dimensional analysis for root cause analysis at scale
- Delegated credentials: Improving the security of TLS certificates
- 2019 @Scale Conference recap
- Simon Marlow, Simon Peyton Jones, and Satnam Singh win Most Influential ICFP Paper Award
- Video @Scale 2019 recap
- Fighting fraud using partially blind signatures
- Systems @Scale 2019 New York recap
- Scribe: Transporting petabytes per hour via a distributed, buffered queueing system
- Releasing a new benchmark and data set for evaluating neural code search models
- Hydra: A framework that simplifies development of complex applications
- MaRS: How Facebook keeps maps current and accurate
- Integrating autoconversion: Facebook’s path from Zawgyi to Unicode
- Networking @Scale 2019 recap
- Register now for @Scale 2019!
- Creating a data set and a challenge for deepfakes
- Powered by AI: Oculus Insight
- Redesigning our systems to provide more control over Off-Facebook activity
- Zoncolan: How Facebook uses static analysis to detect and prevent security issues
- New advances in natural language processing
- A new model for word embeddings that are resilient to misspellings
- Performance @Scale 2019 recap
- Systems @Scale 2019 recap
- Michael F. Cohen awarded 2019 Steven A. Coons award
- EGG: A toolkit for multi-agent language emergence simulations
- Mapping roads through deep learning and weakly supervised training
- Hermes: An open source JavaScript engine optimized for mobile apps, starting with React Native
- Facebook, Carnegie Mellon build first AI that beats pros in 6-player poker
- OIL+VCache: File abstraction for distributed systems
- DLRM: An advanced, open source deep learning recommendation model
- Neural Code Search: ML-based code search using natural language queries
- Facebook’s open source browser contributions
- Aria Presto: Making table scan more efficient
- Delos: Simple, flexible storage for the Facebook control plane
- Efficient, reliable cluster management at scale with Twine
- Building Facebook’s service encryption infrastructure
- Extending DHCPLB: The path from load balancer to server
- Releasing Pythia for vision and language multimodal AI models
- Facebook joins new Urban Computing Foundation
- F8 2019 Day 2 keynote and session videos
- Open source at F8 2019
- F8 2019 Day 1 keynote and session videos
- Open-sourcing F14 for faster, more memory-efficient hash tables
- F8 2019 speakers and sessions now available
- Faster input events with Facebook’s first browser API contribution
- Improving scene understanding through panoptic segmentation
- Aroma: Using machine learning for code recommendation
- Facebook tech news roundup
- Facebook Engineering Fair, London recap
- Open-sourcing Habana back end for Glow
- PyTorch-BigGraph: Faster embeddings of extremely large graphs
- Turing Award presented to Yann LeCun, Geoffrey Hinton, and Yoshua Bengio
- Facebook’s Mechanism Design for Social Good research winners
- Open-sourcing Python Test Runner for multiple tests in parallel
- Building lifelike virtual avatars to help people connect in VR
- Facebook announces next-generation Open Rack frame
- Reinventing Facebook’s data center network
- Accelerating Facebook’s infrastructure with application-specific hardware
- Sharing a common form factor for accelerator modules
- AI blog roundup
- Using aluminum to overcome capacity challenges in submarine cable systems
- Building backbone network infrastructure
- Open-sourcing homomorphic hashing to secure update propagation
- F8 2019 registration now open
- Enabling better global connectivity through new partnerships and technologies
- Open-sourcing Magma to extend mobile networks
- Releasing Hooks for React 16.8
- Open-sourcing SPARTA to make abstract interpretation easy
- FAIR open-sources new ELF OpenGo data set, research, and insights
- Engineering manager Mark Harman wins 2019 IEEE Harlan D. Mills award
- Team behind Infer wins POPL 2019 Most Influential Paper Award
- Opening our newest data center in Los Lunas, New Mexico
- Cross-lingual pretraining sets new state of the art for natural language understanding
- Teaching AI to learn speech the way children do
- No training required: Exploring random encoders for sentence classification
- A new predictive model for more accurate electrical grid mapping
- Zero-shot transfer across 93 languages: Open-sourcing enhanced LASER library
- Facebook open-sources Spectrum 1.0.0 for better mobile image production
- Rethinking data center design for Singapore
- AI year in review
- Open source year in review
- Data centers year in review
2018 97
- Open sourcing wav2letter++, the fastest state-of-the-art speech system, and flashlight, an ML library going native
- DNS over TLS: Encrypting DNS end-to-end
- Nevergrad: An open source tool for derivative-free optimization
- 5 ways Facebook improved compression at scale with Zstandard
- Open-sourcing DeepFocus, an AI-powered system for more realistic VR images
- Open-sourcing PyText for faster NLP development
- HyperLogLog in Presto: A significantly faster way to handle cardinality estimation
- Facebook contributes to MLPerf, open-sources Mask R-CNN2Go
- PyTorch developer ecosystem expands, 1.0 stable release now available
- Léon Bottou and Olivier Bousquet win Test of Time award
- Mobile @Scale — Tel Aviv recap
- FAIR turns five: What we’ve accomplished and where we’re headed
- Using AI and satellite imagery for disaster insights
- Neural Egg Separation: Training ML systems to extract audio/video from noisy/cluttered environments
- New fastMRI open source AI research tools from Facebook and NYU School of Medicine
- Predictive test selection: A more efficient way to ensure reliability of code changes
- Toward a deeper understanding of the way AI agents see things
- Facebook London and U.K. Year of Engineering campaign focus on future of engineering
- Data @Scale – Boston recap
- Making floating point math highly efficient for AI hardware
- Open-sourcing FBGEMM for state-of-the-art server-side inference
- Open-sourcing foundational tools for AI performance
- Getafix: How Facebook tools learn to fix bugs automatically
- Zero-shot learning: Using text to more accurately identify images
- React Conf recap: Hooks, Suspense, and Concurrent Rendering
- Horizon: The first open source reinforcement learning platform for large-scale products and services
- Facebook open-sources new suite of Linux kernel components and tools
- QNNPACK: Open source library for optimized mobile deep learning
- Facebook, NYU expand available languages for natural language understanding systems
- Significantly faster generation and training for AI-based audio systems
- MobileLab: Highly accurate testing to prevent mobile performance regressions
- TIP Summit 2018: Advancing efforts to improve global connectivity
- Improving AI language understanding by combining multiple word representations
- Oculus Connect 5: Tech talks roundup
- Facebook releases Safety Check crisis response tool for Workplace
- Managing data store locality at scale with Akkio
- Open-sourcing StateService: Automating recovery of third-party services after a major outage
- Under the hood: Facebook Marketplace powered by artificial intelligence
- Facebook accelerates AI development with new partners and production capabilities for PyTorch 1.0
- Training AI agents to solve unfamiliar tasks
- Data distillation makes omni-supervised learning possible
- Open-sourcing LogDevice: A distributed data store for sequential data
- 2018 @Scale Conference recap
- Efficient tuning of online systems using Bayesian optimization
- Glow: A community-driven approach to AI infrastructure
- Finding and fixing software bugs automatically with SapFix and Sapienz
- Expanding automatic machine translation to more languages
- Rosetta: Understanding text in images and videos with machine learning
- Scaling neural machine translation to bigger data sets with faster training and inference
- FBOSS: Building switch software at scale and in the open
- Unsupervised machine translation: A novel approach to provide fast, accurate translations for more languages
- Express Wi-Fi Technology Partner program: Getting more people online with fast and affordable Wi-Fi
- Facebook and NYU School of Medicine launch research collaboration to improve MRI
- Open-sourcing Facebook’s Solar-Powered Aircraft Design Tools
- Announcing tools to help partners improve connectivity
- Deploying TLS 1.3 at scale with Fizz, a performant open source TLS library
- Systems @Scale 2018 recap
- Open-sourcing oomd, a new approach to handling OOMs
- Location-Aware Distribution: Configuring servers at scale
- Facebook AI Research Expands With New Academic Collaborations
- XARs: An efficient system for self-contained executables
- Facebook researchers win Test of Time award
- Talk the Walk: Teaching AI systems to navigate New York through language
- Spiral: Self-tuning services via real-time machine learning
- High altitude connectivity: The next chapter
- Migrating Messenger storage to optimize performance
- 2018 Networking @Scale recap
- Accelerate large-scale applications with BOLT
- Building data science teams to have an impact at scale
- Open-sourcing Flipper: a new extensible debugging tool
- How IPv6 deployment is growing in U.S. and other countries
- StatePoint Liquid Cooling system: A new, more efficient way to cool a data center
- Open-sourcing Katran, a scalable network load balancer
- Scaling the Facebook backbone through Zero Touch Provisioning
- Embodied Question Answering: A goal-driven approach to autonomous agents
- AI at F8 2018: Open frameworks and responsible development
- Announcing PyTorch 1.0 for both research and production
- Sapienz: Intelligent automated software testing at scale
- Advancing state-of-the-art image recognition with deep learning on hashtags
- React 360 replaces React VR for streamlined development focus
- ONNX expansion speeds AI development
- Facebook video adds AV1 support
- AV1 beats x264 and libvpx-vp9 in practical use case
- The next step in Facebook’s AI hardware infrastructure
- Fabric Aggregator: A flexible solution to our traffic demand
- Performance @Scale 2018 recap
- Profilo: Understanding app performance in the wild
- Quality assessment of 360 video view sessions
- HotSwap: Bringing hot code reloading to Buck
- Global connectivity: Working together to bring more people online
- Under the hood: Suicide prevention tools powered by AI
- F8 2018 registration now open
- How production engineers support global events on Facebook
- Android @Scale 2018 recap
- Improving Android video on News Feed with Litho
- Under the hood: Multilingual embeddings
- Android native library merging
2017 80
- 2017 Year in review: Software across the stack
- 2017 Year in review: Building immersive experiences
- 2017 Year in review: Better global networks
- 2017 Year in review: Data centers
- ONNX V1 released
- Facebook and Airbus Working Together to Advance High Altitude Connectivity
- Open/R: Open routing for modern networks
- Rethinking Android app compilation with Buck
- Aquila: What’s next for high-altitude connectivity?
- Open-sourcing Sections: Declarative data handling for Litho lists
- Open-sourcing RacerD: Fast static race detection at scale
- Mobile @Scale 2017 recap
- Multithreaded rendering on Android with Litho and Infer
- React 16: A look inside an API-compatible rewrite of our frontend UI library
- Relicensing the GraphQL specification
- Migrating a database from InnoDB to MyRocks
- Relicensing React, Jest, Flow, and Immutable.js
- SLAM: Bringing art to life through technology
- Announcing Yarn 1.0
- Finding inter-procedural bugs at scale with Infer static analyzer
- Rapid release at massive scale
- LogDevice: a distributed data store for logs
- Optimizing 360 photos at scale
- FCR: Open source command runner for network devices
- Steering oceans of content to the world
- Explaining React’s license
- Transitioning entirely to neural machine translation
- Using AI for new visual storytelling techniques in VR
- Building open source tools for Adobe Creative Cloud updates
- Introducing ELF: An extensive, lightweight and flexible platform for game research
- Accelerating growth through Facebook’s Rotational Engineering Program
- Aquila’s successful second flight: Another step forward in bringing the world closer together
- Data @Scale 2017 Recap
- Deal or no deal? Training AI bots to negotiate
- Accelerating machine learning for computer vision
- Dev Tools @Scale 2017 Recap
- Managing resources for large-scale testing
- Jupiter: A high-performance job-matching service
- AL: A new declarative language for detecting bugs with Infer
- Under the hood: Photo Search
- Networking @Scale 2017 Recap
- ParlAI: A new software platform for dialog research
- A novel approach to neural machine translation
- OnlineSchemaChange rebuilt in Python
- Expanded fastText library now fits on smaller-memory devices
- Building Express Backbone: Facebook’s new long-haul network
- Surround 360: Beyond Stereo 360 Cameras
- Facebook’s new tech at F8 2017
- Announcing 360 Capture SDK
- Enhancing high-resolution 360 streaming with view prediction
- Connectivity: A building block approach
- Facebook open source at F8 2017
- Open-sourcing Litho, a declarative UI framework for Android
- Relay Modern: Simpler, faster, more extensible
- Building virtual reality experiences on the web with React VR
- Faiss: A library for efficient similarity search
- ARIA Grid: Supporting nonvisual layout and keyboard traversal
- Evaluating boosted decision trees for billions of users
- Video @Scale 2017 recap
- Hardware Analytics and Lifecycle Optimization (HALO) at Facebook
- OCP Summit 2017 — Facebook news recap
- The end-to-end refresh of our server hardware fleet
- Introducing Big Basin: Our next-generation AI hardware
- Introducing Bryce Canyon: Our next-generation storage platform
- Designing 100G optical connections
- The software ecosystem for storage hardware
- Airtel and BCS, with support from Facebook, to build shared fiber backhaul connectivity in Uganda
- End-to-end optimizations for dynamic streaming
- Spatial audio — bringing realistic sound to 360 video
- Machine Learning @Scale 2017 recap
- Using Apache Spark for large-scale language model training
- Register now for F8 2017!
- Beringei: A high-performance time series storage engine
- Building scalable systems to understand content
- Disaggregate: Networking recap
- Building Zero protocol for fast, secure mobile connections
- This browser tweak saved 60% of requests to Facebook
- The growing ecosystem around open networking hardware
- Sustainable materials in the data center
- Legacy support on IPv6-only infra
2016 108
- Favorite hacks of 2016
- Facebook Open Source 2016 year in review
- Reviewing Aquila’s first full-scale test flight
- Made in NY: The engineering behind social recommendations
- Yoga: A cross-platform layout engine
- Artificial intelligence, revealed
- Keyframes: Delivering scalable, high-quality animations to mobile clients
- Spam Fighting @Scale 2016
- Open population datasets and open challenges
- Facebook demonstrates record-breaking data rate using millimeter-wave technology
- Introducing Backpack: Our second-generation modular open switch
- Accelerating innovation and powering new experiences with AI
- Delivering real-time AI in the palm of your hand
- Introducing Community Cellular Manager: A management and deployment suite for small-scale cellular networks
- An open approach for switching, routing, and transport
- Networking @Scale Boston
- Continuous MySQL backup validation: Restoring backups
- Components for Android: A declarative framework for efficient UIs
- Building an efficient neural language model over a billion words
- Client-side ranking to more efficiently show people stories in feed
- A comparison of state-of-the-art graph processing systems
- Wedge 100: More open and versatile than ever
- Android accessibility debugging with Stetho
- Building one of the highest-capacity subsea cables in the Pacific
- Yarn: A new package manager for JavaScript
- Lessons from building social experiences in VR
- React reaches 50,000 stars
- Redesigning the HHVM JIT compiler for better performance
- Facebook in Los Lunas: Our newest data center
- DHCPLB: An open source load balancer
- Scalable and secure access with SSH
- Peter O’Hearn elected Fellow of the Royal Academy of Engineering
- Highlights from @Scale 2016
- Facebook announces new tech at @Scale 2016
- BrowserLab: Automated regression detection for the web
- Apache Spark @Scale: A 60 TB+ production use case
- 360 video stabilization: A new algorithm for smoother 360 video viewing
- MyRocks: A space- and write-optimized MySQL database
- Smaller and faster data compression with Zstandard
- Facebook opens lab to others to validate infrastructure software
- Segmenting and refining images with SharpMask
- Robotron: Top-down network management at scale
- FAIR open-sources fastText
- Inside Facebook’s hardware labs: Moving faster with more collaboration
- Surround 360 is now open source
- Flying Aquila: Early lessons from the first full-scale test flight and the path ahead
- Harnessing light for wireless communications
- What’s new in Facebook open source
- The mobile device lab at the Prineville data center
- Making Facebook self-healing: Automating proactive rack maintenance
- Registration is open for @Scale 2016
- Introducing OpenCellular: An open source wireless access platform
- Building a better way to write posts in multiple languages
- Primitive sets for Android
- Memory optimization for feeds on Android
- Lighting the way to deep machine learning
- A path to unsupervised learning through adversarial networks
- Developing the technical language behind creativity for Canvas
- Growing the Wedge/Wedge 100 community
- Data @Scale, June 2016 — Recap
- Improving CSS quality at Facebook and beyond
- Safety Check: Streamlining deployment around the world
- Introducing DeepText: Facebook’s text understanding engine
- Python in production engineering
- Networking @Scale, May 2016 — Recap
- Introducing “Here and Now” — a new short film shot with Facebook Surround 360
- Facebook Seattle moves into Dexter Station
- Introducing Open/R — a new modular routing platform
- Artificial intelligence, revealed
- Introducing FBLearner Flow: Facebook’s AI backbone
- Automatically push commits to GitHub with FBShipIt
- Wangle — an asynchronous C++ networking and RPC library
- Facebook Chef cookbooks
- Automatic memory leak detection on iOS
- Building and managing iOS model objects with Remodel
- Powering Facebook experiences with AI
- Introducing Facebook’s new terrestrial connectivity systems — Terragraph and Project ARIES
- React Native: A year in review
- Open-sourcing ReDex: Making Android apps smaller and faster
- Introducing Facebook Surround 360: An open, high-quality 3D-360 video capture system
- Breaking ground at Clonee
- Under the hood: Building accessibility tools for the visually impaired on Facebook
- Engineering for nostalgia: Building a personalized “On This Day” experience
- Mobile @Scale London recap
- Dive into React Native performance
- Dragon: A distributed graph query engine
- Facebook Boston tech talk: GraphQL
- Open Compute Project U.S. Summit 2016 — Facebook news recap
- Facebook’s new front-end server design delivers on performance without sucking up power
- Opening designs for 6-pack and Wedge 100
- Introducing Lightning: A flexible NVMe JBOF
- OpenBMC: One board management software for all hardware at Facebook
- How we built Facebook Lite for every Android phone and network
- Connectivity Lab custom-designs two-axis gimbal for air-to-air and air-to-ground laser communications
- Performance @Scale 2016 recap
- Facebook open sources rich text editor framework Draft.js
- Facebook AI Research launches partnership program
- Register for OCP U.S. Summit
- Adopting an open approach to global networks with the Telecom Infra Project
- Connecting the world with better maps
- Gear VR to get Dynamic Streaming for 360 video
- NetNORAD: Troubleshooting networks via end-to-end probing
- In pursuit of secure open source software
- Under the hood: Shipping #friendsday videos
- Hardware and firmware attacks: Defending, detecting, and responding
- Telecom industry embraces Open Compute Project
- Facebook in Ireland: Our newest data center
- Next-generation video encoding techniques for 360 video and VR
2015 96
- Favorite hacks of 2015
- Open source in 2015: A year of growth
- Improving software RAID with a write-ahead log
- Why we chose to move to HTML5 video
- Facebook to open-source AI hardware design
- Continuing to build News Feed for all types of connections
- HHVM support for PHP 7
- Under the hood: Broadcasting live video to millions
- Making React Native apps accessible
- HIKE: Our quick, simple accessibility primer
- Optimizing Facebook for iOS start time
- Open networking advances with Wedge and FBOSS
- Security @Scale 2015: Engineering Security
- Building the Presto community
- Teaching machines to see and understand: Advances in AI research
- Milestones and plans for the OCP Networking Project
- Building open source communities for security
- Building for emerging markets: The story behind 2G Tuesdays
- Performance instrumentation for Android apps
- More momentum on OCP adoption
- Under the hood: Building 360 video
- Improving the Linux kernel with upstream contributions
- Optimizing Android bytecode with ReDex
- Instrumenting meetings at Facebook
- Inside @Scale 2015
- @Scale 2015 – Facebook announcement recap
- GraphQL: A data query language
- Relay: Declarative data for React applications
- IPv6: It’s time to get on board
- React Native for Android: How we built the first cross-platform React Native app
- Facebook Seattle’s five-year anniversary: Celebrating a culture of entrepreneurialism
- How we use it: Building Out the Window with React Native and Parse
- Reducing FOOMs in the Facebook iOS app
- Writing code that writes code — with Hack Codegen
- Under the hood: Place Tips at Lollapalooza
- Inside the 2015 Edge Conf
- OpenBMC for server: Porting and supporting new features for “Yosemite”
- Gearing up for @Scale 2015
- The Parse SDK: What’s inside?
- The technology behind preview photos
- Improving Facebook’s performance on Android with FlatBuffers
- Building communications networks in the stratosphere
- Linux IPv6 improvement: Routing cache on demand
- OSCON 2015: How Facebook open-sources at scale
- Using ISC Kea DHCP in our data centers
- Osquery: Introducing query packs
- Under the hood: Box’s HHVM migration
- Sustainability @Scale 2015 recap
- Announcing XHP-JS: Building efficient user interface components with Hack, React, and XHP
- Facebook in Fort Worth: Our newest data center
- 72 hours to launch Celebrate Pride
- Fighting spam with Haskell
- Delivering high scroll performance
- Building “add a link”
- Building Nuclide, a unified developer experience
- Inside Data@Scale 2015
- Futures for C++11 at Facebook
- Oculus scientist wins computer vision award
- Under the hood: Building Moments
- Open-sourcing Facebook Infer: Identify bugs before you ship
- Inside the HHVM lockdown
- Meet the wogrammers — women in engineering
- Embracing open source security
- Recommending items to more than a billion people
- Facebook expands AI research team to Paris
- Spam Fighting @Scale Recap
- Under the hood: Facebook’s cold storage system
- How RocksDB is used in osquery
- Parse now running MongoDB on RocksDB
- HHVM adoption news
- Web performance: Cache efficiency exercise
- Updating Our Open Source Patent Grant
- Mobile performance: Tooling infrastructure at Facebook
- Introducing the Accessibility Toolkit
- F8: Big technology bets and open source announcements
- Introducing Fresco: A new image library for Android
- Classes: Performance and network segmentation on Android
- React Native: Bringing modern web techniques to mobile
- Introducing ComponentKit: Functional and declarative UI on iOS
- Augmented Traffic Control: A tool to simulate network conditions
- Even faster: Data at the speed of Presto ORC
- Open Compute Project U.S. Summit 2015 – Facebook News Recap
- Introducing “Yosemite”: the first open source modular chassis for high-powered microservers
- Facebook Open Switching System (“FBOSS”) and Wedge in the open
- Introducing “OpenBMC”: an open software framework for next-generation system management
- Serving Facebook Multifeed: Efficiency, performance gains through redesign
- Shimmer for Android
- Introducing Origami Live
- Stetho: A new debugging platform for Android
- Networking @Scale Recap
- Introducing “6-pack”: the first open hardware modular switch
- Fast Rendering News Feed on Android
- Faster Photos in Facebook for iOS
- FAIR open sources deep-learning modules for Torch
- Alibaba joins WebScaleSQL
- Inside Wikipedia’s transition to HHVM
2014 62
- Favorite Hacks of 2014
- The art and science behind auto-enhance for iOS
- 12 Days of Open Source
- Data @Scale – Boston Recap
- Flow, a new static type checker for JavaScript
- Solving the Mystery of Link Imbalance: A Metastable Failure State at Scale
- Introducing data center fabric, the next-generation Facebook data center network
- Announcing the Hack Transpiler
- Audience Insights query engine: In-memory integer store for social analytics
- Year class: A classification system for Android
- Security @Scale 2014 Recap
- Introducing Proxygen, Facebook’s C++ HTTP framework
- Making News Feed nearly 50% faster on iOS
- Introducing osquery
- Engineering Culture: Code ownership
- @Scale 2014: Recap of Data Track
- Introducing AsyncDisplayKit: For smooth and responsive apps on iOS
- Building Mobile-First Infrastructure for Messenger
- Announcing the WEST (Women Entering and Staying in Tech) Mentorship Pilot
- @Scale 2014: Recap of Web Track
- @Scale 2014: Recap of Mobile Track
- Facebook, configuration management, community, and open source
- Lessons from Deploying MySQL GTID at Scale
- Introducing TODO: Working together to make open source easier
- Introducing mcrouter: A memcached protocol router for scaling memcached deployments
- Sustainability @ Scale Recap
- Debugging file corruption on iOS
- Making Facebook’s software infrastructure more energy efficient with Autoscale
- #bewhoyoucansee: Larry Schrof
- #bewhoyoucansee: Ming Hua
- 9.9 million lines of code and still moving fast – Facebook open source in 2014
- Improving Facebook on Android
- Introducing “Wedge” and “FBOSS,” the next steps toward a disaggregated network
- F8 Developer Conference – Hacker Way Recap
- Open-sourcing Haxl, a library for Haskell
- HydraBase – The evolution of HBase@Facebook
- Saving capacity with HDFS RAID
- Announcing the 2014 Facebook Grace Hopper Scholarship
- Introducing Pop, the animation engine behind Paper
- Technical Program Managers at Facebook
- Scaling the Facebook data warehouse to 300 PB
- Hack Developer Day Recap
- Large-scale graph partitioning with Apache Giraph
- Under the Hood: warp, a fast C and C++ preprocessor
- Announcing The Connectivity Lab at Facebook
- WebScaleSQL: A collaboration to build upon the MySQL upstream
- Hack: a new programming language for HHVM
- Introducing the Facebook Messenger for Android beta testing program
- Looking back on “Look Back” videos
- Building Paper
- Building and open-sourcing Bolts, a mobile developer tools library
- Facebook’s 4th Annual Hacker Cup Finals
- HHVM – Implementing MySQLi
- An analysis of Facebook photo caching
- Under the Hood: Building and open-sourcing flint
- An analysis of Facebook photo caching
- Under the Hood: Building and open-sourcing fbthrift
- A Decade of Building Facebook
- Introducing Conceal: Efficient storage encryption for Android
- Under The Hood: Facebook Accessibility
- Airlock – Facebook’s mobile A/B testing framework
- Scaling Mercurial at Facebook
2013 40
- 2013: A Year of Open Source at Facebook
- Under the Hood: Building and open-sourcing the Rebound animation library for Android
- Under the Hood: Building and open-sourcing RocksDB
- Building a faster Messenger
- Facebook Open Academy: Bringing open source to CS curricula
- The Mature Optimization Handbook
- Presto: Interacting with petabytes of data at Facebook
- Under the Hood: Building posts search
- Under the hood: MySQL Pool Scanner (MPS)
- Update on the Facebook for Android beta testing program
- Flashcache at Facebook: From 2010 to 2013 and beyond
- Scaling Apache Giraph to a trillion edges
- Secure browsing by default
- Windex: Automation for database provisioning
- Introducing the Facebook for Android Beta Testing Program
- Q&A: Joining the engineering team in London
- TAO: The power of the graph
- Wormhole pub/sub system: Moving data through space and time
- Adding :face: to every IP: Celebrating IPv6’s one-year anniversary
- Watchman: Faster builds with large source trees
- Speeding up the Activity social plugin
- Speaking a whole new language: DConf 2013 at Facebook
- Buck: How we build Android apps at Facebook
- Facebook Celebrates Women in Computing with Grace Hopper Scholarship
- Under the Hood: The natural language interface of Graph Search
- Scaling memcache at Facebook
- LinkBench: A database benchmark for the social graph
- Hacker Cup 2013
- Three Optimization Tips for C++
- Under the Hood: Indexing and ranking in Graph Search
- User Experience Lab: How we designed a new News Feed using your feedback
- Under the Hood: Building out the infrastructure for Graph Search
- McDipper: A key-value cache for Flash storage
- Under the Hood: Dalvik patch for Facebook for Android
- Like Box plugin: Now 2-4x faster
- Preparing for a technical interview with programming contests
- Celebrating a year of fighting email phishing with DMARC
- Fighting spam with pure functions
- Under the Hood: Building Graph Search Beta
- Under the Hood: Automated backups
2012 62
- Under the Hood: Rebuilding Facebook for Android
- Under the Hood: The JavaScript SDK – Truly Asynchronous Loading
- Under the hood: Facebook Messenger for Firefox
- Speeding up PHP-based development with HipHop VM
- Happy Hacktober
- Under the Hood: The JavaScript SDK – The use of polyfills
- Facebook at the espnW Hack Day
- Under the Hood: Integrating standalone mobile app features into Facebook for iOS
- Under the Hood: Scheduling MapReduce jobs more efficiently with Corona
- Under the Hood: The JavaScript SDK – error handling
- Product engineering at Facebook
- 2012 Grace Hopper Conference Recap
- Interning at Facebook: Building the new mobile advertising interface
- Under the Hood: Building the App Center recommendation engine
- Building the WordPress plugin
- Timed releases for mobile apps
- Monitoring cache with Claspin
- Optimizing for the MacBook Pro with Retina display
- A faster, better link shim
- Under the Hood: Building Facebook Camera
- Under the hood: Rebuilding Facebook for iOS
- Under the hood: Improving messages
- Announcing the Facebook Grace Hopper Scholarship
- Building and testing at Facebook
- Ship early and ship twice as often
- Announcing Facebook Engineering in London
- Get that job at Facebook
- Hackathon: Building a better Events view
- Under the Hood: Timeline apps behind Facebook engineering
- Under the Hood: Hadoop Distributed Filesystem reliability with Namenode and Avatarnode
- Under the Hood: Network implementation for World IPv6 Launch
- Folly: The Facebook Open Source Library
- Stay focused and keep hacking
- Back-end build systems and toolchain: Adam Simpkins
- Three tips to succeed as a programmer
- Release engineering and push karma: Chuck Rossi
- Moving Fast and Staying Reliable: Mary Pimenova
- From intern to Photos team full-timer: Jen Dolson
- A hack of epic proportions: Building a QR code on the roof
- Girls Programming Camp: Beginning a career in tech
- Under the Hood: Improving Facebook Photos
- The 2012 Facebook Hacker Cup World Champion is crowned
- Dare2BDigital: Why STEM needs women
- Under the Hood: Building the Location API
- Under the Hood: Building Facebook Messenger for Windows
- Research at Facebook: “Nothing Is Riskier than not Taking Risks”
- MySQL and Database Engineering: Mark Callaghan
- From News Feed to Timeline: Q&A with Raylene Yung
- Announcing the 2012-2013 Facebook Fellows
- Meet a Facebook Engineer: Ben Billings
- Meet a Facebook Engineer: Ben Gertzfield
- DMARC: Building Open-Source Email Authentication Technologies
- Meet a Facebook Engineer: Philip Su
- Meet a Facebook Engineer: Andrew (Boz) Bosworth
- My First Year Fighting Spam
- Meet a Facebook Engineer: Josh Wiseman
- Meet a Facebook Engineer: Vijaye Raji
- Under the Hood: Mobile Timeline
- Meet a Facebook Engineer: Peter Wilson
- Meet a Facebook Engineer: Keith Adams
- Building Timeline: Scaling up to hold your life story
- Announcing Facebook’s 2012 Hacker Cup
2011 54
- Meet a Facebook Engineer: Haiping Zhao
- Meet a Facebook Engineer: Carlos Bueno
- Meet a Facebook Fellow: Marek Olszewski
- The HipHop Virtual Machine
- High School Hackathon
- Meet a Facebook Fellow: Adrienne Porter Felt
- Compassion Research Day
- Celebrating Women in Tech at Grace Hopper 2011
- Making HPHPi Faster
- Jocelyn Goldfein at WITI: Women are the Power Users of Social
- Watir to WebDriver: Unit Test Frameworks
- Software Design Glossary
- Making Facebook Self-Healing
- Interning at Facebook: Who Goes Where When (and Why It Matters)
- Keeping the Site Reliable While Moving Fast
- Interning at Facebook: Answering Your Questions on Quora
- Interning at Facebook: Failing Harder
- Celebrating Facebook Seattle’s First Anniversary
- Building Facebook Messenger
- Interning at Facebook: Bridging Marketing and Engineering
- Interning at Facebook: Coding, Inspiration and BBQs
- Moving an Elephant: Large Scale Hadoop Data Migration at Facebook
- Doppler: Internet Radar
- Building Video Calling
- My Experience with Facebook’s Grace Hopper Women in Computing Scholarship
- Reflections on the Open Compute Summit
- Exciting Results from World IPv6 Day
- Facebook Helps Celebrate Women in Computing
- Facebook and World IPv6 Day
- Visualizing Facebook’s PHP Codebase
- How Project Triforce Prepared our Software Stack for Prineville
- Introducing Login Approvals
- Working with Students to Improve Indexing in Apache Hive
- Building the Send Button
- Inside Facebook Messages’ Application Server
- Hackamonth: Mixing Things Up
- Thoughts on Software Quality
- Designing a Very Efficient Data Center
- Scaling the Messages Application Back End
- Inside the Open Compute Project Server
- Building Efficient Data Centers with the Open Compute Project
- One Mobile Site to Serve Thousands of Phones
- HipHop for PHP: More Optimizations for Efficient Servers
- Accelerating Network Innovation with the Open Networking Foundation
- Building Realtime Insights
- Facebook Hacker Cup Finals: A Champion is Crowned
- Hybrid Incremental MySQL Backups
- Hackathon 22: Redesigning Pixelcloud
- Developing Facebook’s New Photo Viewer
- Live Commenting: Behind the Scenes
- HTML5 Games 0.1: Speedy Sprites
- Supercell: test infrastructure for any open source project
- World IPv6 Day: Solving the IP Address Chicken-and-Egg Challenge
- Scalable memory allocation using jemalloc
2010 32
- A New Year of Facebook Fellowships
- Join Optimization in Apache Hive
- Visualizing Friendships
- Looking at the code behind our three uses of Apache Hadoop
- Announcing the Facebook 2011 Hacker Cup
- The Full Stack, Part I
- The Underlying Technology of Messages
- New Cooling Strategies for Greater Data Center Energy Efficiency
- Google Summer of Code 2010
- Optimizing Data Center Energy Usage
- Crowdsourcing Mobile Device Capabilities
- Using HTML5 Today
- More Details on Today’s Outage
- HipHop for PHP: six months later
- Reflecting on 3 years at Facebook
- Scaling Facebook to 500 Million Users and Beyond
- Internet Cartography
- Announcing the Grace Hopper Scholarship
- BigPipe: Pipelining web pages for high performance
- Protecting Privacy with Referrers
- The Life of a Typeahead Query
- Touchscreens are Fingerpaint
- Facebook engineering seeks new friends in Seattle
- Intro to Facebook Search
- Working with Zuck
- Making Facebook 2x Faster
- Internal Company Dashboard
- XHP: A New Way to Write PHP
- Site Reliability Engineering at Facebook
- Facebook becomes a USENIX Patron
- HipHop for PHP: Move Fast
- Bootcamp: Growing Culture at Facebook
2009 22
- Real-World Web Application Benchmarking
- Facebook Engineering Bootcamp
- Building a Better Photo Uploader
- Inviting you all to Hackathon
- Names in translation and social language modeling
- Reducing markup size
- Every Millisecond Counts
- Facebook on TopCoder
- Designing the Facebook username land rush
- 10th Annual System Administrator Appreciation Day
- A New Look at the Path to Popularity
- Hammering Usernames
- Chat reaches 1 billion messages sent per day
- Hive – A Petabyte Scale Data Warehouse using Hadoop
- The Facebook Puzzle Conciseness Contest
- Needle in a haystack: efficient storage of billions of photos
- Adapting Open Source Software
- Scaling Facebook Live Feed for the Inauguration
- XHProfLive: Continuous function-level performance stats from production
- The way Facebook is made
- Chat Stability and Scalability
- Behind the scenes with the new Facebook features in Apple’s iPhoto ’09
2008 11
- Scaling memcached at Facebook
- How to Hack Out an Election Counter
- Facebook’s Scribe technology now open source
- The All-Night Hackathon Is Back!
- 10 billion photos
- Cassandra – A structured storage system on a P2P Network
- Scaling Out
- Thrift: (slightly more than) one year later
- Hadoop
- Facebook Chat
- Welcome to the Facebook Engineering Blog!