HN user
Use Luna. It's more performant than 5.5 and it's cheap. Hopefully it's cheap because it's more environmentally friendly than the bigger models. So you're doing a good thing. If it's a smaller model it may even be faster, but I haven't looked into it yet.
Decent sandbox + sandbox override experience with pi coding agent... pi-sandbox uses the same sandbox tech that claude code uses, although it uses a fork that's a little behind, and I'm not sure exactly why it uses a fork.
You can install pi, then install pi-sandbox locked to the current version. Here it is described how pi-sandbox plus an additional extension allow you to have the experience where a sandbox is used, but you can fall back to unsandboxed with approval required. https://github.com/carderne/pi-sandbox/issues/50
Regarding the photos, the "storage scope" feature on Graphene should make that a non-issue. Microphone would be rough, but at least you would see a "microphone icon" show up on the top of the phone if an app was actually listening.
Here's why I use Cursor. My company pays for it, although I could switch to Claude Code or use Codex more since I also have ChatGPT enterprise account.
* Perhaps could be solved with the right terminal software, but I like the GUI for seeing my running agents and viewing all my conversations
* Works with multiple model providers in the same tool. I probably worry about cost optimization more than my employer would care for me to, but I frequently switch between openai/anthropic and switch between model sizes to use the tool that I think can get the job done for the least money. Another thing I like is having a long conversation with an expensive model, then I can switch to 5.4-nano to cheaply extract some little piece of information or summary from the conversation. Really this is big being able to switch model providers throughout the months without having to change my interface.
* Good support for the various ways of providing context. Rules, AGENTs.MD/CLAUDE.md files (if you want it to automatically read those), skills. Good hook support.
* I think the agent diff review experience is pretty good, but maybe it works similarly when you hook the cli agents into an editor, IDK.
* The default shell sandbox behavior is quite good. Every shell command runs in some sort of sandbox so that read only commands work without approval. The model asks for more permissions when it tries to do something that needs more permissions like network access or writing outside of the workspace directory. I know Claude code has a similar feature you can use.
* Good fork / revert conversation to checkpoints, with the option of reverting the code or just reverting the conversation.
* Feels decent that I am an API customer through Cursor. I don't hit Claude limits. Cursor doesn't have an incentive to limit reasoning or token usage, although they do have an opposite incentive.
* They are reasonably responsive to bugs and feature requests through their forum.
* Works well with a lot of repos / folders added to your workspace. I probably should organize all my stuff under a single directory, but alas I have like 8 different folders added to my workspace and it handles this well. Perhaps Claude --add-dir support works fine too.
DOWNSIDES:
* They are not quickly adding the best open source models to Cursor. Like Kimi 2.6 or whatever. Possibly not incentivized to given their Composer models.
* Don't love the subagent support. I can define custom subagents although it is not easy to get models to use mine instead of the builtin ones. The builtin ones do not allow me to control what model they run, so they will always run something like composer-2-fast, which is a fine model for all I know, but I would like to control it. Also, I would like if you could optionally make the subagent experience more first class. Like browse all the subagents and continue conversations with them or switch their model etc, although that is probably tricky / weird.
The article does say
No human in the loop
If true, it's quite irresponsible. They are admitting to allowing a agent to autonomously execute code on the network. Autonomously perform hacking activities.
For directed coding (implementing an already specified plan) or asking questions about a codebase I use 5.3 codex with medium reasoning effort. It is relatively quick feeling.
I like Sonnet 4.6 a lot too at medium reasoning effort, but at least in Cursor it is sometimes quite slow because it will start "thinking" for a long time.
I've been using 5.1-codex-max with low reasoning (in Cursor fwiw) recently and it feels like a nice speed while still being effective. Might be worth a shot.
Wow, that's pretty epic and satisfying
Beautifully done!
We spend ~$20,000 per month in AWS for the product I work on. In the average day we do not launch an EC2 instance. We do not do any dynamic scaling. However, there are many scenarios (especially during outages and such) that it would be critical for us to be able to launch a new instance (and or stop/start an existing instance.)
FYI, you can get a pixel 8a for $200 or less if you don't care about condition. Support end date in 2031.
I run with the dangerous option on my work computer. At first I was thinking I would be good if I just regularly kept full disk backups. But my company at least pays lip service to the fact that we want to protect our intellectual property. Plus I think it might be irresponsible to allow an AI model full internet access unsupervised.
So now I use a docker compose setup where I install Claude and run it in a container. I map source code volumes into the container. It uses a different container with dnsmasq with an allowlist.
I initially wanted to do HTTP proxying instead of DNS filtering since it would be more secure, but it was quite hard to set it up satisfactorily.
Running CLI programs with the dangerous full permissions is a lot more comfortable and fast, so I'm quite satisfied.
No the comment meant that aider and Claude code are CLI programs, so if you can run a terminal in your niche editor, then you are good to go
It's pretty easy if you just use the MCP Python library. You just put an annotation on a function and there's your tool. I was able to do it and it works great without me knowing anything about MCP. Maybe it's a different story if you actually need to know the protocol and implement more for yourself
Firefox has a "reader view" which works well for things like that.
Hi, I appreciate you sharing. I've been starting to use this advice with a different tool. Just FYI, this sentence kind of came out of nowhere and it wasn't clear what you meant: > The foundational LLM models right now are what I'd estimate to be at circa 45% accuracy and require frequent steering
Do your rules count as frequent steering and lead to increased 'accuracy', or is that the 'accuracy' you're seeing with your current workflow, rules and all?
I've also heard that THCA breaks down into THC over time, so maybe there's a chance that when it's shipped it's compliant with the Hemp bill, but some time later if the cops seize and test it, then you actually now have illegal cannabis in your possession. Not sure to what extent that is true, but seems reasonable.
Doesn't help that the post didn't define their definition of devops, and then they went on to use it in a variety of contexts, seemingly contradictory. As someone who is a software engineer, but doesn't necessarily keep up on the latest hype, this post made me head spin when I tried to read it casually.
I recently had GPT 3.5 write me a function for translating the number under point to an ISO timestamp. It handled whether the epoch number was milliseconds or seconds since epoch. I kinda had to iterate on it and ask it to do some things different, but it never wrote invalid code.
What interface are you using? I've so far only used ChatGPT through the web interface. (I pay for plus). I just ask it for things that are standalone snippet-like things. Perfect functions on the standard library, or an emacs lisp snippet for converting a unix epoch to an ISO timestamp. I'm not sure I could allow it to read my company's code, but is that what you're doing when you're using as part of a large project like that?
No it's not possible in a newer LTS (> 20.04, don't have laptop on hand) Ubuntu KDE. There's no scroll speed configuration, as sibling comments discuss. Switching driver to libinput or whatever made it configurable, but other bugs popped up, and suddenly every accidental palm touch sent my cursor clicking somewhere.
Yes, I also really enjoyed this piece. It has a nice and loose style... just throwing ideas out without trying to "prove" them, but nonetheless presenting interesting ideas and likely some useful mental models.
Donating money to projects that you use heavily doesn't sound like any kind of traditional paid model to me.
The author's linked page has a horizontal scroll, which seems non-ideal to me https://karl-voit.at/tags/lazyblorg/
Do you sit on a cushion or no?
You also have the monitor set up so that it's hanging over you somewhat? When you rest the laptop on your chest, do you use the laptop display? Isn't that a weird position to type? What do you mean suspend the keyboard in front of you?
haha, you kinda made it sound like your setup was super minimalist, but none of your descriptions are making sense in my mind
I'm also a big fan of working while reclined. I am less fancy, and probably less ergonomic about it. I will sit on a recliner couch, or in the warm season I also sit on a zero gravity chair[0] outside, on my deck, in the shade. In both instances I sit with a pillow behind my back... to make the seat less deep / keep me propped up a bit more I guess?
FWIW, I probably sit 2-4 hours reclined on couch or recliner chair, 2 hours standing, 2 hours in traditional chair.
0: https://external-content.duckduckgo.com/iu/?u=https%3A%2F%2F...
Something sounds wrong, but I don't know exactly what you mean by "basic template"? Like a Spring boot example app?
People are talking about REPL development, and they are right, but one of the best things that you can do with CL that you can't do with other REPLs, such as Clojure, is... at least using Slime in Emacs, probably other editors too... when there's an error/exception or whatever, you drop right into the stack at the error location. You can see all the values of the variables. You can evaluate expressions. Travel up the stack and do the same things. That is SO powerful, and can save so much time debugging simple errors.
No it's not a degradation. It puts Ctrl in a sane place much easier to reach with your pinky. And if you really were going to use your thumb on the left control, then it's about the same distance to the Caps Lock key. I'm replying to the original asker with my advice, if you want to read more about what I think :)