Tag: 工程实践 (6 articles)

Hugging Face publishes an AI Agent glossary to clarify confusing and rapidly evolving terminology, providing developers with a clear mental model.

An expert critiques current AI agents for being too 'human'—lacking rigor, patience, and focus, and tending to compromise when faced with difficulties, revealing fundamental flaws in their design.

Meta shares its post-quantum cryptography migration framework, introduces PQC Migration Levels, and emphasizes the need to act now to defend against 'store now, decrypt later' attacks.

Meta successfully migrated 50+ use cases from a long-maintained internal WebRTC fork to a modular architecture synced with the latest upstream, solving the industry-wide 'forking trap' problem through a dual-stack architecture.

LangChain proposes a 6-point checklist before building agent evaluations, emphasizing manual analysis of 20-50 real failure traces before automating tests.

LangChain shares its core philosophy for building AI agent evaluation systems: more evals aren't better; instead, precisely define and measure the agent behaviors you care about to guide its evolution.