eschluntz

I did a double take at your name. Hello doppelganger!

Very cool! I've designed a lot of Hilbert Curve marble tracks, using OpenSCAD and python

Yeah if there's more interest I can publish the GH and onshape! I just need to sanity check that I didn't leave any keys in it or anything...

Thanks!

Blog post about my mechanical design, robot math, stress testing, and competition for "Mech Warfare", a small airsoft-only battlebots competition this summer

https://github.com/eschluntz/eschluntz.github.io/blob/gh-pag...

copy and pasted into a spreadsheet from the website, then claude wrote a little script to convert it into js

<no commitment that this file will stay up to date, use at your own risk, etc etc>

How else m I going to geotarget you with ads? /s

I thought about this, but I've seen several times where the signs say that the train is running late, but it's actually there and leaves perfectly on time!

The schedule has been much more reliable since the electrification

Thanks :)

bullet trains should be displayed in red, but I haven't tested it yet at the right time and I've been too lazy to write tests with mocked time / gps :)

"22nd st" shows up separately

Also there's no backend - everything is in localStorage

Made this while procrastinating some actual work, in an effort to get some more dopamine from crossing off todos.

Completing an item gets you a confetti explosion, a satisfying ding, and shakes the item.

Section headers (starting with "#") are even more exciting to cross off

hi! You can do /cost at any time to see what the current session has cost

It is API billing like AWS - you pay for what you use. Every time you exit a session we print the cost, and in the middle of a session you can do /cost to see your cost so far that session!

You can track costs in a few ways and set spend limits to avoid surprises: https://docs.anthropic.com/en/docs/agents-and-tools/claude-c...

>> npm install -g @anthropic-ai/claude-code

>> claude

Thanks! We all dogfood Claude every day to do our own work here, and solving our own pain points is more exciting to us than abstract benchmarks.

Getting things done require a lot of booksmarts, but also a lot of "street smarts" - knowing when to answer quickly, when to double back, etc

Right now no, but if you run in docker, you can use `--dangerously-skip-permissions`

Some commands could be totally fine in one context, but bad in a different i.e. pushing to master

hi! I've been using Claude Code in a very complementary way to my IDE, and one of the reasons we chose the terminal is because you can open it up inside whichever IDE you want!

Part of our vision is that because Claude Code is just in the terminal, you can bring it into any IDE (or server) you want! Obviously that has tradeoffs of not having a full GUI of the IDE though

We are definitely aware of this (and working on it for the web UI), and that's why Claude Code goes directly through the API!

hi! I've been working on demos where I let Claude Code run for hours at a time on a sandboxed project: https://x.com/ErikSchluntz/status/1894104265817284770

TLDR: asking claude to speed up my code once 1.8x'd perf, but putting it in a loop telling it to make it faster for 2 hours led to a 500x speedup!

Even when I feel this, 90% of any novel thing I'm doing is still old gruntwork, and Claude lets me speed through that and focus all my attention on the interesting 10% (disclaimer: I'm at Anthropic)

Totally agree! I wrote this a few years ago: https://blog.cobaltrobotics.com/you-cant-read-code-that-you-...

As someone who's not a frontend developer, this was definitely faster for me than writing it without any AI, although it did still take an hour or two to get the exact details and literal corner cases correct.

From doing a lot of coding with AI I've built up a pretty good intuition of what's faster to do by hand (changing some hyperparameters, or fixing a 1 line bug) and what's faster for the AI.

In this case, I started in 2D and added complexity bit by bit. Being able to say "Now make it 3D" and have that just mostly work in 10s was pretty wild.

Oops, forgot I had made the repo originally private. Fixed now!

A re-creation of the classic Windows 3D "Pipes" screensaver, written entirely by Claude 3.5 Sonnet over many iterations!

(full disclosure - I work at Anthropic)

This was an experiment to try to generate a medium sized (240 line) program without writing a single line of code myself, only with prompting.

I had the best luck starting simple and gradually adding features (including starting 2D). Claude was able to debug its code about 2/3 of the time when it created errors, and 1/3 of the time I had to open up the code and help it more explicitly.

I documented some of my prompts and the code here: https://github.com/eschluntz/pipes-screensaver/blob/main/REA...

It says 5x the free tier usage, but I agree we could be more explicit about what that is. Just made a ticket :)

Oh no! I use Apple voice control into Claude and then copy and paste code back and forth between my IDE and Claude mostly.

For longer writing,I really like going for a walk for 45 minutes and brain dumping on a topic, and transcribing it. Then I write a brief outline and have Claude fill it out into a document, explicitly only using language that I used in the transcript. Then edit via voice

(I'm biased as I work for anthropic)

I broke my hand 2 months ago and Claude 3.5 sonnet has been writing all my code for me. It's been awesome