News for hackers. Stuff that matters.

atworkc

Re:Agents don't need memory. They need documentation on Sunday October 04, 2026 @04:05AM (#49951704)
I've settled on just a simple folder called `workbench` for some reason the new models know exactly what's up with it. Git ignored

The only "prompt" is in AGENTS.md saying that this thing exists and there's a map.md <- which is a one liner reference to whatever the agent stores in there.

And usually, I tackle a new feature, and at some point tell it to store to jot down notes in workbench if I'm comfortable with it (and if it needs to be stored in memory)

This makes it easy to just spawn other agents and such from a good point, I just point them that stuff is in workbench.

Token costs, seem decent and I can always just delete stuff in there as its purpose is ephemeral.

Re:Ask HN: How do you as a normal swe use AI? on Monday July 20, 2026 @03:49PM (#48983971)
It does feel like an organic process, like I have to intuitively feel the solution for the need I have, can't really use all the shiny things effectively if I don't feel the pain in my bones which they solve.

I've been moving my goldilocks zone a bit towards scheduled tasks simply because I've actually felt the need and pain they solve, more like I have some backburner things I want to take care of but they're so unimportant that I'm unwilling to spend more than 5m here and there thinking about them so I add a scheduled taks every other day which does some leg work, I review for 5 minutes and then forget about it until the next iteration.

Re:Ask HN: How do you as a normal swe use AI? on Monday July 20, 2026 @02:30AM (#48974977)
Definitely appreciate the write-up!

And I think I'm in a somewhat similar place with how I use the tools as yourself.

How do you feel about your workflow, do you feel you're underusing the tooling or overusing it?

Ask HN: How do you as a normal swe use AI?

So am looking for more normal people, experienced people using AI, but I feel like whenever I hit youtube or other platforms the AI culture is way deep in the trenches and don't feel like this is how the majority of people tackle work, especially at their jobs.What I see is mostly solo-devs going full battlestar gallactica on things. And I just want to see other people, normal people talk about...

( Read More... | 9 comments | 2 points | on HN )
Servers / Computers your device is currently communicating with, e.g. github servers when you load the link (well probably a cdn edge one)
Re:Claude 4 on Friday May 23, 2025 @05:05AM (#44071205)
"AI does badly on my test much like other AI's did before it, therefore I don't immediately see much improvement" is a fair assumption.
```

Attach a probability for the answer you give for this e.g. (Answer: x , Probability: x%)

Question: How many times does the word “blue” appear in the following sentence: “The sky was blue and my blue was blue.”

```

Quite accurate with this prompt that makes it attach a probability, probably even more accurate if the probability is prompted first.

Maybe a tad philosophical/pedantic but many mathematicians follow the "Mathematical realism" approach and would say that any new mathematics be it simply describing existing shapes is actually a form of discovery in the world of mathematics
I don't think the author is suggesting we should all cram it alongside each other. These "view modes" could be incorporated via hotkeys or such.

Am thinking about 3D software where switching view modes is fundamental to building things in them. Hmm

Show HN: NPM package for tagging logs with an emoji based on an identifier

https://www.npmjs.com/package/mojilogger

Hey guys! I just released a NPM package to use for debugging seas of console logs. Basically, I had lots of use cases where I have multiple instances of similar logs that I logged with ids and such and they were always hard to trace when all of them were happening at the same time. So i built this library to help with visually locating log sources.

( Read More... | no comments | on HN )