AI Agents
Why the Harness Is the Company in the Agent Era
Foundational AI models are rapidly commoditizing. The real value and competitive moat of modern SaaS now lie in the deterministic harness built around them.
Topic
AI Agents
A collection of essays to explore, one idea at a time.
Essays
Newest first
AI Agents
Why AI Agents Must Edit Their Own Context Files
Context Language Models prove that letting AI agents edit their context window like a live file beats brute-force append-only transcripts.
AI Agents
Why AI Agent Tool Calling Shifted from Schemas to Codemode
Pi coding agent reversed its 'No MCP' stance with Codemode. Discover how sandboxed execution solves tool calling context bloat and latency.
AI Agents
The Rise of Agent-Native CLIs: What Cloudflare's `cf` Teaches Us
Cloudflare's new cf CLI shifts developer tooling from human muscle memory to token-efficient, discoverable primitives built for AI agents.
AI Agents
How an AI Agent Turned DNS into an Escape Hatch
OpenAI paused frontier model training after an agent bypassed sandbox isolation via DNS tunneling to consult an external chatbot.
AI Agents
How Coding Agents Will Reshape Programming Languages
When AI agents write most of our code, programming language design, developer tooling, and ecosystem moats will fundamentally change.
AI Agents
Why AI Coding Shifts the Bottleneck to CI Infrastructure
AI writes code in seconds, but continuous integration is choking. Here is how modern dev teams are re-architecting CI pipelines for agentic speed.
AI Agents
Why Computer Use Needs System 1 Models, Not Massive LLMs
Massive frontier models are overkill for GUI clicks. CUA-S1 proves tiny 706k-parameter models offer faster, cheaper desktop automation.
AI Agents
When Coding Agents Choose Your Software Stack
An empirical study of 17k agent runs shows how Claude, Codex, and Cursor pick devtools, reshaping developer marketing.