Meta

Modernizing the Meta Ads Service With an Open-Source Kernel Scheduler

Linux kernel upgrades risked introducing latency regressions across Meta's ad serving fleet, which operates at scale where milliseconds of latency degradation significantly impact ads performance.

observability distributed-systems
5 min
Meta

10 Years of Meta’s Commitment to Python

Meta needed to maintain and advance Python as a critical language across its engineering infrastructure while supporting the broader open-source community that sustains the language.

general
5 min
Meta

How Meta Engineered Ultra-Narrow Batteries for AI Glasses

How to design and manufacture ultra-narrow batteries that fit within the temple arms of smart glasses while providing sufficient power for cameras, speakers, AI workloads, and displays.

general
5 min
Meta

Capacity Efficiency at Meta: How Unified AI Agents Optimize Performance at Hyperscale

Meta needed to automatically identify and remediate performance inefficiencies across their massive infrastructure to reduce power consumption and free up engineering capacity.

observability distributed-systems
5 min
Meta

Escaping the Fork: How Meta Modernized WebRTC Across 50+ Use Cases

Meta needed to modernize WebRTC across 50+ use cases while maintaining synchronization with upstream open-source development, avoiding the drift that typically occurs when large projects fork internally.

distributed-systems real-time-systems
5 min
Meta

Trust But Canary: Configuration Safety at Scale

Safely deploying configuration changes at scale while minimizing the risk of widespread failures caused by faulty configurations.

observability distributed-systems
5 min
Meta

AI for American-Produced Cement and Concrete

Designing high-quality, sustainable concrete mixes that are produced in the United States while optimizing for performance characteristics.

ml-systems general
5 min
Meta

KernelEvolve: How Meta’s Ranking Engineer Agent Optimizes AI Infrastructure

Meta needed to automatically optimize low-level infrastructure and kernel-level parameters for AI ranking models to improve performance without manual tuning.

ml-systems distributed-systems
5 min