Agensh: Microsoft Ran 1,024 Agents with No Boss and Won
Microsoft Research open-sourced Agensh, a multi-agent harness that runs 1,024 coding agents without a central orchestrator—and gets better benchmark results than single-agent systems.
Block’s Dessa Drops Text From the Voice AI Pipeline
Block’s Dessa is a speech AI trained only on human conversation, with no text or ...
AAIF’s Four New Open Agent Projects: What to Use Now
The Agentic AI Foundation added Agent Skills, Open Responses, OpenEnv, and OpenSandbox this week. Two ...
AI Has Erased 200,000 Creative Jobs. What the Data Says
The U.S. lost 200,000 creative jobs since ChatGPT launched. Film down 27%, publishing down 70,000. ...
Rakazo: Self-Host the Open-Source Grok Bot Alternative
Rakazo is an open-source, self-hosted alternative to Grok Bot, OpenAI Dots, and Meta Muse. Bring ...
MLflow 3.9: AI Assistant and MemAlign for Agent Developers
MLflow 3.9 ships a Claude Code-powered AI assistant, MemAlign judge optimizer (30-50% accuracy gain), and a pre-built agent quality dashboard. Here is what ...
TorchNeuron: Run PyTorch on AWS Trainium in One Line
Amazon's TorchNeuron lets existing PyTorch code run on Trainium2 and Trainium3 with one line change. ...
Cursor Computer Use: Agents Now Test Their Own Code
Cursor’s coding agents can now control your computer. On October 23, Anysphere shipped Computer Use ...
Harness AI Budgets: Your AI Cost Policy May Be Failing You
73% of engineering teams have an AI cost policy. Only 13% can verify it works. ...
1Password $500M ARR: AI Agents Drive 500% Growth in Non-Human Identities
1Password crossed $500M ARR, but the real signal is 500% growth in non-human identity users. ...