OpenKnowledge launches as an open-source, AI-native knowledge base from Inkeep — 347 HN upvotes, real production AI infra behind it, and a business model worth understanding.
Recall adds fully local, session-persistent memory to Claude Code. 82 HN upvotes on launch confirm real developer pain — but stale memory may be worse than none.
Z.ai's GLM-5.2 scores near GPT-5.5 on benchmarks while hallucinating 3× less—new data shows why brute-force AI scaling breaks production reliability.
Loupe exposes a structural flaw in AI-assisted development: when an agent writes both code and tests, green CI proves consistency, not correctness.
A Google employee built a working travel-to-calendar app in 2 hours with no code. Here's what that really means for developers and security teams.
A community audit of 347 flagged Dev.to posts exposes why generic NLP classifiers fail on technical writing — and what platform teams must fix before shipping.
AI hallucinations aren't a trust problem — they're a design problem. Here's how GraphRAG and MCP reshape what AI can reliably do in production codebases.
Number 0 Labs releases Iroh 1.0, a Rust-based networking library that simplifies peer-to-peer data sync, hole-punching NAT traversal, and distributed app development.
Kobo's rejection of valid ePub files exposes how Adobe's DRM validation creates a de facto proprietary standard, breaking interoperability promises.
Salesforce announced a $3.6 billion deal to acquire Fin, the AI-powered customer support platform formerly known as Intercom. Here's what developers need to know.