Andrej Karpathy has a concise way to puncture AI hype: “When you get a demo and something works 90% of the time, that’s just the first nine.” For enterprise teams turning large language models into real products, that line is… Read More »From Demos to Dependable: Engineering the ‘March of Nines’ for Enterprise AI Agents
Scaling from a clever LLM demo to a reliable, long‑running AI agent in production is less about squeezing a few more percentage points out of the underlying model and more about how you “harness” that model. That’s the core argument… Read More »Why Better LLMs Aren’t Enough: LangChain’s Harrison Chase on Harness Engineering and Deep Agents
As large language models (LLMs) get more capable, many teams assume that upgrading to the latest model will automatically make their AI agents production-ready. Harrison Chase, co-founder and CEO of LangChain, argues the opposite: model quality is necessary but no… Read More »Why Harness Engineering, Not Just Better Models, Will Decide Which AI Agents Reach Production
The command line is re-emerging as a central interface for AI agents — and now Google Workspace is moving firmly into that world. A new open source tool, googleworkspace/cli, brings Gmail, Docs, Sheets, Drive, Calendar and other Workspace APIs into… Read More »Google Workspace CLI Turns Gmail, Docs, and Sheets Into a Unified Interface for AI Agents
OpenAI has spent years pushing the frontier of large language models. Internally, though, one of its most consequential AI deployments isn’t a new model at all. It’s a data agent, built by just two engineers in three months—70% of its… Read More »Inside OpenAI’s In‑House AI Data Agent: How Two Engineers Built a Tool 4,000 Employees Use Daily
Generative AI coding tools are often pitched as smart assistants: handy for scaffolding features or unblocking you on unfamiliar APIs, but not something you’d trust with a production system. One recent experiment with Google AI Studio and Gemini 3.0 Pro… Read More »From Jam Session to Discipline: What Vibe Coding with Google AI Studio Teaches About Managing Overeager AI Coders
Block, the fintech company led by former Twitter co-founder Jack Dorsey and parent to Square, Cash App, Tidal, and the open source AI orchestration project Goose, has executed one of the largest workforce reductions the sector has seen: more than… Read More »Inside Block’s 40% Layoffs: How Agentic AI Is Reshaping Fintech Organizations
ServiceNow is putting hard numbers behind a vision many enterprises are still only piloting. Inside its own organization, the company says 90% of employee IT requests are now resolved autonomously, and those cases are closed 99% faster than when handled… Read More »ServiceNow’s Autonomous Workforce: From AI Assistants to Governed Virtual Employees
AT&T’s AI platform was quietly burning through roughly 8 billion tokens every day. For most enterprises, that level of usage is a sign that adoption is working. For Andy Markus, AT&T’s chief data officer, it was also a warning: pushing… Read More »Inside AT&T’s Agentic AI Stack: How 8 Billion Tokens a Day Led to a 90% Cost Cut
Anthropic is positioning its new Claude Cowork platform as the missing piece between experimental AI agents and dependable enterprise-grade automation. At its virtual “Briefing: Enterprise Agents” event, the company argued that 2025’s rush into AI agents mostly yielded stalled pilots… Read More »Anthropic’s Claude Cowork Aims to Turn Enterprise AI Pilots Into Production Workhorses
Anthropic is taking its fast-growing coding agent off the desktop and into your pocket. Claude Code, which has surged in popularity among both developers and non-technical users, now has a mobile control layer called Remote Control that lets you steer… Read More »Anthropic’s Claude Code Goes Mobile: Inside the New Remote Control Mode
When a single engineer can stand up a production SaaS interface in about an hour, the question for most engineering leaders is no longer, “Is agentic coding fast enough?” It is, instead, “What kind of governance makes that speed survivable… Read More »How Treasure Data Built Governance So One Engineer Could Ship a Production AI Tool in an Hour