Don't be a meat proxy

Don't be a meat proxy

Aug 03, 2026

Too often<br>I ask a question in Slack or<br>leave feedback under a merge/pull request or<br>argue with friends in a WhatsApp group and<br>get back:

Claude said: [giant response verbatim]

Please don't do this.<br>I mean, I've done this.<br>But I've been on the receiving end too many times now.<br>This is not adding value.<br>I can talk to Claude myself.<br>It's going to be faster and I get to control the context.<br>I don't need a meat proxy in between.

Reading AI output is extra effort.<br>It's verbose,<br>frequently contains all too plausible nonsense,<br>and is increasingly jargon dense.<br>I recently got this sentence from Claude:

NATS control-plane events: stream leader election / R3 quorum re-form during pod churn.

Jesus. I had to lookup almost every word to make sense of this.

By all means, prompt AI.<br>But don't just relay the output.<br>Read it, understand it, validate it, and then write a response in your own words<br>(a decent certificate that you've done the prior steps).<br>Making that effort is value you can add.

Take code review in particular.<br>Shipping some code can be done with close to zero effort now:<br>Copy/paste the ticket description into Claude Code.<br>Don't look at the code or read what Claude has written.<br>If there's any feedback from reviewers,<br>copy/paste that into Claude Code as well.<br>If necessary, iterate.

That works.<br>But who has done the implementation?<br>The reviewers did, using Claude Code, and you as a meat proxy.