IDEAS, DECISIONS & LESSONS
Notes on AI, architecture and building useful products.
7 articles
26 Sep 2026 · 7 min read
agents
Cursor's new Rollouts and Security Reviewer agents watch your PR from merge to production and file revert PRs on their own. I like the idea. I don't trust the default settings, and here's the policy I'd put around them before turning either one loose.
Read more14 Sep 2026 · 5 min read
agents
On September 3, 2026, Cursor let you swap its own cloud-agent infrastructure for eight other backends, including Vercel Sandbox. That's not a partnership announcement. It's Cursor telling you where its moat actually is.
Read more05 Sep 2026 · 4 min read
agentic coding
Cursor shipped Origin — a waitlist-only Git forge that lets cloud agents push code without ever touching GitHub. A technical lead's read on what it means when 35% of merged PRs are already agent-authored.
Read more25 Aug 2026 · 4 min read
agent architecture
I already run a self-hosted autonomous engineer. When I wanted Cursor's cloud agents too, I didn't pick one — I made the executor swappable. Here's the integration architecture that turns any hosted coding agent into a first-class engine behind my own surface, in four small pieces.
Read more22 Jul 2026 · 7 min read
ai
OpenCode #1, Cursor #2, Claude Code #3 — but which AI coding tool is right for your team? A practical Tech Lead framework with 5 evaluation criteria, real trade-offs, and a recommendation matrix for startups vs enterprises.
Read more30 Mar 2026 · 6 min read
agentic-ai
Cursor's March 2026 release brings parallel subagents, BugBot Autofix, and cloud agents — a genuine shift from AI coding assistant to AI workflow engine. Here's what it means for real teams.
Read more26 Feb 2026 · 13 min read
ai
What is AI hallucination? Why does AI invent facts? Learn detailed strategies for minimizing hallucinations in Claude, Gemini, Copilot, and Cursor — from prompt engineering to RAG, grounding, and verification loops.
Read more
