Tag: developer-tools

All the articles with the tag "developer-tools".

• Using REA to build my own meeting notebook

  Published: 13 Oct, 2026

  • 10 min read

  

  I used REA to investigate Granola, then built a meeting notebook for myself. What helped, what broke, and what the experiment taught me.

• Why the dark factory is a Michelin kitchen

  Published: 11 Oct, 2026

  • 12 min read

  

  In February I said the dark factory wasn't the point. Aim for lights-off anyway, but run it like a Michelin kitchen: sample the dishes, fix the kitchen when the same mistake shows up twice, then franchise.

• Running an organization of agents through Tightbeam

  Published: 14 Sep, 2026

  • 19 min read

  

  Six weeks running forty-three coding agents through Tightbeam: what it records, how reviews and rules hold, what broke, and who should try it.

• For every agent that writes code, two agents review it

  Published: 2 Aug, 2026

  • 8 min read

  Six weeks of agent-written code with two reviewer agents per writer. What the gate caught: leaked secrets, silent side effects, and what it costs.

• This is the feedback loop you didn't know you needed

  Published: 28 Jul, 2026

  • 8 min read

  A week-plus with Lavish, the local review surface for agent-generated HTML: how the annotate-to-agent loop works, when the wait pays off, what it costs.

• How I actually run a swarm of coding agents

  Published: 13 Jul, 2026

  • 8 min read

  

  tmux can't tell you an agent is blocked or done. Herdr surfaces each agent's real state, and lets the agents drive the multiplexer themselves. My actual setup.

• Four Claude Code Commands for Managing Context

  Published: 9 Jul, 2026

  • 5 min read

  Branch, fork, background, and btw each send Claude Code work somewhere else, keeping side questions, experiments, and long runs out of one giant session.

• Trying Bumblebee

  Published: 27 May, 2026

  • 6 min read

  

  Running Perplexity's read-only endpoint scanner: zero catalogue findings, sixty-three MCP servers I never installed, and one freshness pattern worth knowing.

• Getting More Out of Claude Code With Remember and Superpowers

  Published: 28 Apr, 2026

  • 5 min read

  

  These two plugins made Claude Code easier to use for real engineering work by tightening the loop around context, planning, execution, and review.

• Experimenting with the Entire CLI for AI Code Attribution

  Published: 3 Mar, 2026

  • 5 min read

  A two-run case study of Entire CLI attribution: what the first test left ambiguous, what the prompt-only rerun clarified, and what still needs interpretation.