When doing this remove write permissions on the test file, it will do a much better job of staying the course over long periods. I've been doing this for over a year now.

Well said. I have to review PRs of non-software developers nowadays.

The “what is this trying to do?” has never been harder to answer than before. It creates scenarios where 99% is correct, but the most important area is subtly broken. I prefer it to be human, where 60-80% will be correct, and the problematic areas begin to smell more and more gradually.

In my experience LLMs, at times, may hide the truth from you in a haystack made of needles.

This very matches my observation. The error isn't due to incorrect code—it's code that looks specific to your system but is actually generic patterns applied from the training process. The structure is correct, the logic is sound, it just doesn't interact with what your source code actually does.

Harder to catch because nothing is factually wrong. You have to ask: could this output have been produced without actually reading my codebase?

I carefully review far more than 14k LoC a week… I’m sure many here do. Certainly the language you write in will greatly bloat those numbers though, and Node in particular can be fairly boilerplate heavy.

I tried this for a week and gave up. Required far too much back and forth. Ate too many tokens, and required too much human in the loop.

For this reason I don’t think it’s actually a good name. It should be called planning-shit instead. Since that’s seemingly 80%+ of what I did while interacting with this tool. And when it came to getting things done, I didn’t need this at all, and the plans were just alright.

Been using Elixir with agents for over a year. Seemed like an obvious choice to me.

Node is great, but scaling Elixir threads is more so.

“NPM registry directory” sounds terribly confusing in my opinion, reads like a grammatical mishap

then maybe "npm registry search"?

Especially because they mention unmaintained software (volta) in the same breadth as maintained software (fnm/mise)

Same, I am looking forward to starting my own company in the next year or two, and exploring the CTO role, especially after so many years as a Staff+ operating as an arm of CTOs and learning the ropes.

Check out the earlier posts in the series. They’re probably more relevant for that stage. What I’ve learned over the years is that the CTO role varies a lot depending on the company and personal context. I just tried to share my journey.

You wouldn't use it in an offline-first application. But you can use it an offline-first application, and it isn't even very difficult.

I have built electron apps for fun that utilize React Server Components entirely offline.

The official blog post disagrees.

> Even if your app does not implement any React Server Function endpoints it may still be vulnerable if your app supports React Server Components.

Oops, yeah. My bad, you're right. It makes sense that any flight server that is capable of interpreting server functions would be vulnerable whether the codebase used them or not. It's an issue in the transport mechanism and not the actual RPC implementation.