Posted inBlog
What Makes an Agent Swarm Actually Work
For most of the past two years, multi-agent work meant one orchestrator handing tasks to sub-agents — a delegation pattern, not a swarm. Something changed recently with the OpenAI Astra…
Posted inBlog
When AI Stops Advising and Starts Delivering
The shift in how knowledge workers relate to AI assistants has been gradual but unmistakable. For years, the dominant pattern was retrieval and advice: ask a question, get a summary,…
Posted inBlog
Beyond Context Engineering: The Scaffolding AI Agents Actually Need
The scaffolding matters as much as the model. That has become clear as teams move from simple prompt chains to multi-step autonomous agents. Around 2023, the conversation focused on prompt…
Posted inNews
The CCA Exam as a Field Guide: Six Anti-Patterns Every Agentic Engineer Should Know
For most of the past few years, building an AI agent meant wiring a model to a few tools and hoping the prompt held. The architecture was simple because the…
Posted inBlog
The Jagged Frontier Was a Measurement Error — Draft: The Surgeon-in-the-Dark Read
Shaping the flow of ideas at the frontier of cognitive systems. --- Cursor's coding assistant — a tool built to help programmers write software — recently solved Problem 6 of…
Posted inBlog
The Copy-Paste Problem: How a Structured Task Record Changes Multi-Agent Handoffs
For a long time, the way most of us worked with multiple AI tools was embarrassingly manual. Run a task in one tool, copy the output into a message, paste…
Posted inBlog
When AIs Discover Management
Shaping the flow of ideas at the frontier of cognitive systems. --- Sixteen AI agents built a working C compiler in two weeks. The compiler processes 100,000 lines of Rust…
Posted inBlog
Delegation or Coordination? The AI Agent Choice That’s Harder to Reverse Than You Think
Something shifted for me when two major AI agent releases dropped within twenty minutes of each other earlier this year. The question had been framed as a capability race —…
Posted inBlog
The Jagged Frontier Was a Measurement Error
For the past three years, the phrase "jagged frontier" has been one of the most useful mental models in AI strategy. The basic idea: AI is surprisingly good at some…
Posted inBlog
The Agent Economy Is Already Here. Nobody’s Governing It.
Part 2 of the OpenClaw story was expected to bring more security coverage — more CVEs, more enterprise hand-wringing, maybe a regulatory response. What emerged instead was something stranger and,…