techxiv
A tale of two parameter architectures—and how we unified them 
Accelerating on-device ML on Meta’s family of apps with ExecuTorch 
Fluid: How we built serverless servers - Model Context Protocol (MCP) explained: An FAQ
Resolving digital threats 100x faster with OpenAI - Policy Zones: How Meta enforces purpose limitation at scale in batch processing systems
Accelerating LinkedIn Sales Navigator's search system with Spark transformations - How Meta keeps its AI hardware reliable
- Grep a million GitHub repositories via MCP
- ChatGPT agent System Card
How Go 1.24’s Swiss Tables saved us hundreds of gigabytes - How we tracked down a Go 1.24 memory regression across hundreds of pods
- Invideo AI uses OpenAI models to create videos 10x faster
- Using AI to make lower-carbon, faster-curing concrete
Notion’s hosted MCP server: an inside look - Rolling out Santa without freezing productivity: Tips from securing Figma’s fleet
A Higgs-bugson in the Linux Kernel - An inside look at Meta’s transition from C to Rust on mobile
- Introducing Northguard and Xinfra: scalable log storage at LinkedIn
- Canvas, meet code: Building Figma’s code layers