AI agent failures often stem from poor context management rather than model limitations. Static context loads fixed information every session, while dynamic context retrieves data on demand, balancing…
Anysphere, the company behind the Cursor code editor, has trained its own frontier AI model called Composer 2.5 on Qwen 2.5 Coder using novel reinforcement learning techniques, enabling it to compete …
Developers can now run Zhipu AI's GLM 5.2 model inside Anthropic's Claude Code coding environment by routing API calls through OpenRouter, using the ANTHROPIC_BASE_URL environment variable to point to…
Zhipu AI released GLM 5.2, a 744B parameter open-weight mixture-of-experts model that rivals Claude Opus in coding and visual design quality at lower inference cost. The model's MoE architecture activ…
A loop of loops is a design pattern for AI agents that manage recurring work, where a parent loop coordinates multiple child loops to share context and handle handoffs without human intervention. This…
Zhipu AI released GLM 5.2, a 744-billion-parameter mixture-of-experts open-weight model that outperforms Claude Opus on visual design tasks at lower cost. The model features a longer context window, i…
Sakana AI launched Fugu, a multi-agent AI system that orchestrates Claude, GPT, and Gemini through one API to outperform individual frontier models on benchmarks. The system uses a mixture-of-agents a…
ByteDance released Seedance 2.5, an AI video generation model capable of producing 30-second native clips with up to 50 multimodal references. The model addresses key limitations in AI video workflows…
Anthropic's Claude Code, paired with the Gmail MCP server, enables users to build a skill that automatically reads inbox newsletters, scores AI stories, generates a ranked digest, and drafts LinkedIn …
Anthropic's Claude Code introduces five effort levels—Low, Medium, High, Max, and Ultra Code—that control the thinking token budget for reasoning before acting. The /effort command allows users to mat…
Anthropic's Claude Code introduces a /goal command that lets AI agents run until measurable conditions are met, preventing premature stops by defining verifiable finish lines. The command creates an o…
Anthropic's Claude Code now supports /goal and /routines commands that enable developers to define persistent objectives and schedule recurring tasks for autonomous execution, reducing the need for ma…
Claude Code lacks persistent memory across sessions, losing architectural decisions, bug fixes, and project context. Tools like Memarch, Hermes, and GBrain offer solutions for storage, context injecti…
Seventy-six percent of large companies now have a Chief AI Officer, and many more are creating informal versions of the role. A new 4-step career roadmap outlines how employees can become their organi…
OpenAI rolled out live job search and resume tailoring features to ChatGPT in 2025, allowing users to find job listings and customize applications through natural language prompts. The feature pulls l…
Sakana AI's Fugu multi-model orchestration system costs 5x more and runs 4.5x slower than Anthropic's Claude Opus 4.8 while delivering similar results, raising questions about the practical value of m…
ByteDance's Seedance 2.0 now generates native 4K video but costs five times more than 720p, with a five-second clip at 4K costing about $2.50 versus $0.50 at 720p. Generation speed also slows signific…
ByteDance's Seedance 2.5 and Google's Gemini Omni Flash compete in AI video generation, with Seedance excelling in motion realism and facial consistency, while Gemini Omni Flash handles complex scene …
Cursor developer Anysphere is reportedly training a 1.5 trillion parameter large language model from scratch using SpaceX compute infrastructure, signaling a shift from being a model wrapper to a fron…
An agent loop consists of three components: a trigger, an action, and a stop condition. This structure enables autonomous AI agents to run iteratively without human intervention, making decisions at e…