vanpelt

7 ms·

1.
by vanpelt 8mo ago
Ahh, understood. Today you can install the iPad version to your mac, not the best experience but would get you what you want. I'll do some more thinking about a more Mac OS native experience.
2.
by vanpelt 8mo ago
Oooo nice, not really yet... Claude stores your session history in ~/.claude/projects/-Path-to-git-repo on your local machine. I should definitely add a "teleport" command like the one claude itself has which let&
3.
by vanpelt 8mo ago
(1) This one is on the top of my list. I have a branch that sets up proper iOS notifications. Coming soon. (2) Yeah, I've considered making an actual desktop version. A hacky one exists already. You can run `catnip run` from your d
4.
Show HN: Catnip – Run Claude Code from Your iPhone Using GitHub Codespaces(github.com)
9 points by vanpelt 8mo ago|6 comments
5.
by vanpelt 2y ago
Tailwind is especially well suited for LLM generation because all of the styles come inline with the HTML. Could definitely just ask it to write styles but this approach also improves token efficiency.
6.
by vanpelt 4y ago
https://nsjail.dev is a common tool for sandboxing
7.
by vanpelt 4y ago
A100’s are Nvidia GPU’s. You can rent them from providers like AWS or LamdaLabs. The readme has instructions for downloading the original GPT2 weights from OpenAI. You can also train a very simple version on a smaller dataset from your l
8.
A Public Dissection of a PyTorch Training Step(wandb.ai)
1 points by vanpelt 5y ago|0 comments
9.
by vanpelt 5y ago
Pretty cool this post is already the first result in Google’s index. How does that work?
10.
by vanpelt 5y ago
This would be awesome for terraform! It could potentially derive the details of SSL, k8s, Load Balancers, and databases for each cloud from comments
11.
by vanpelt 6y ago
Hey Lenny! Congrats on the launch, so excited to see your product come together.
12.
by vanpelt 6y ago
My bad, this was using non-Accelerated TensorFlow on a 2.3GHz 8-Core i9.
13.
How I made our cloud software self-hostable in a single Docker container(wandb.com)
5 points by vanpelt 7y ago|0 comments
14.
by vanpelt 7y ago
GCP charges ~$0.60/hour for the GPU/CPU/MEM equivalent in colab. If bandwidth is costing 10k per day, how much is the free colab compute costing Google?
15.
by vanpelt 8y ago
Thanks for sharing man, I hope it gives you hope that so many people are engaging with your story. I'm from Iowa so I know how brutal the winters can be, can't imagine what it's like without a roof over your head. Fortune l
16.
by vanpelt 8y ago
What does he mean by GPS? Is he just getting location info from the IP or using the browser geolocation API? I’ve run my fair share of surveys on mTurk and FigureEight. The best way to prevent scams are with honeypots. Always have a que
17.
Starting a Second Machine Learning Company, 10 Years Later(medium.com)
2 points by vanpelt 8y ago|0 comments
18.
by vanpelt 8y ago
Weights & Biases | SF, CA | FT, ONSITE, VISA | www.wandb.com Weights and Biases builds cutting-edge tools for machine learning. Our software makes it easier to build production-ready machine learning models at scale. We help ML teams vi
19.
by vanpelt 14y ago
Just a heads up... If you think D3, crowdsourcing, distributed systems, or the color "baby poop green" is cool; CrowdFlower is hiring engineers: http://crowdflower.com/about/jobs
20.
Finally, a way to prevent goatse...(blog.doloreslabs.com)
1 points by vanpelt 18y ago|0 comments