jerezzprime

I find that the LLM is great for helping with the starting problem. I will go for a walk and basically brain dump my thoughts and ideas about a topic/doc into a voice note, where I can ramble, wander from point to point, etc. Then I transcribe it, get a model to take that and turn it into something cohesive, and then I spend multiple hours reading it, reworking it, and editing it. I am not very good at writing docs and this method produces something with far more detail and quality than I would ever get on my own. But AI;DR right?

To play devil's advocate here, it is possible for someone to simultaneously use AI to write and also not be a meat proxy. I'm not saying that is the norm, but I have written docs where the vast majority of the words come from AI, but I also spent multiple days on the substance and content.

I bet this is the inverse problem. Someone thought "oh a prescription is definitely privacy sensitive information, I should ensure the OS does the right thing" not realizing this counter productive behavior.

What does "100% generated" mean? Do you mean a movie executive told Claude to "make me a 100 million dollar blockbuster, make no mistakes" or do you mean that folks use generative tools to help write the screenplay, generate scenes, etc, or something else?

If you use GitHub Copilot you can switch between them in the same session if you want.

Interestingly, on mobile the first "broken" example, lets me continue to scroll the page when I touch down in the example area. The second "fixed" example traps my scroll. All the things light up, but I can't scroll the page if I touch down in that area.

There is full support for GitHub sandboxes today, and in the near future the remote session experience will improve, so your laptop (or your phone, or the web) can be the control plane.

Yes get a coffee. Being able to execute 5 things at once is amazing, but it's a recipe for burnout. We have to be more careful and explicit about how we spend our time, and that means more explicit time away. If this thing makes you 10x more effective (I truly believe it can), you can afford to spend 20% less time behind the desk and more time doing whatever it is that actually makes you happy. Hopefully your manager understands that calculus.

I'd be interested in seeing actual agent benchmarks (eg CC or Copilot CLI with grep removed and this tool instead).

For example, I have explored RTK and various LSP implementations and find that the models are so heavily RL'd with grep that they do not trust results in other forms and will continually retry or reread, and all token savings are lost because the model does not trust the results of the other tools.

Maybe not safe for valuables. What about stuff that has no value to anyone else? I'm not a villain from Ocean's Eleven, no one is stealing my passwords to break into my elaborate safe.

Let's imagine a scenario. For your entire life, you have been taught to respond to people in a very specific way. Someone will ask you a question via email and you must respond with two or three paragraphs of useful information. Sometimes when the person asks you a question, they give you books that you can use, sometimes they don't.

Now someone sends you an email and asks you to help them fix a bug in Windows 12. What would you tell them?