A developer argues that teams should standardize AI usage by building a 'harness' around the model—tools, context, feedback loops—rather than focusing solely on prompts or models. The key insight is t…
The AI agent development stack is shifting from frameworks to harnesses—control loops that wrap models and manage task decomposition, retries, and context. Leaders from LlamaIndex, LangChain, Anthropi…
SpaceX acquired AI coding startup Cursor for $60 billion, highlighting the booming vibe-coding market. Other startups like Lovable and Replit are raising billions as investors bet on AI-assisted codin…
Agent orchestration spans three distinct layers—expression, runtime, and observability—that are often conflated in debates about multi-agent systems. A June 2025 clash between Cognition's 'Don't Build…
Cognition has developed an automated system to estimate the productivity of its autonomous AI software engineer, Devin, by measuring how many productive engineering hours each session saves. The syste…
Anthropic released Claude Fable 5, its first public Mythos-class model, achieving state-of-the-art results on long, multi-step reasoning and autonomous task completion. The model tops benchmarks like …
Anthropic suspended access to its Claude Fable 5 and Mythos 5 models for all customers after a US government directive citing a potential jailbreak as a national cybersecurity risk, a move Anthropic d…
Artificial Analysis, an independent AI benchmarking platform, launched its Coding Agent Benchmarks and Index at a June 11 event in San Francisco featuring speakers from Cognition, Cursor, and NVIDIA. …
The tech industry shipped four types of memory systems this week, none of which match what users understand the term to mean. Apple paid Google an estimated $1 billion annually for a system-wide seman…
Sarah Guo, a prominent AI investor and Cognition backer, published an essay arguing that the most defensible AI companies will be those focused on "untrainable" work—integrating models into messy, pri…
AI coding agents and development platforms in 2026 have shifted software engineering from manual coding to intent-based development, with tools like Atoms deploying coordinated AI teams to produce ful…
Cognition introduced FrontierCode, a new benchmark that evaluates code on mergeability rather than just unit-test passing, with tasks built by open-source maintainers requiring over 40 hours each. The…
Microsoft launched seven in-house MAI models, including its first reasoning model MAI-Thinking-1, aiming for self-sufficiency from OpenAI. Google released Gemma 4 12B, an open multimodal model for age…
Anthropic, OpenAI, Cognition, Sierra, LangChain, and leading practitioners have published an open standard for building production-grade agentic products, codified as a canonical specification with ru…
Enterprises are increasingly adopting "model routing" to match tasks with the cheapest adequate AI model, a shift that threatens revenue for OpenAI and Anthropic. Glean CEO Arvind Jain estimates 95% o…
Cognition has introduced an AI Productivity Guarantee for its Devin engineering agent, promising enterprise customers up to $10 million in credits if the tool fails to deliver engineering value equal …
Cognition launched Devin Desktop on Tuesday, transforming its Windsurf IDE into an "Agent Command Center" that lets developers manage local and cloud coding agents from a single editor. The new app de…
Anthropic, OpenAI, Cognition, Sierra, LangChain and leading practitioners have jointly developed a canonical standard for building production-grade agentic products, codified in a new open-source repo…
Anthropic surpassed OpenAI to become the world's most valuable AI startup at a $965 billion valuation, driven by enterprise adoption of Claude Code. The proliferation of AI coding agents continued wit…
Cognition CEO Scott Wu said his company's AI coding agent Devin is not designed to replace human programmers, despite the startup's $1 billion funding round at a $26 billion valuation. Wu, a former ch…