thejash

6 ms·

1.
by thejash 16d ago
I was there in the early days (90's, early 2000's) It was categorically, undeniably better--that's where the nostalgia is coming from. There are lots of reasons why this happened (and those are interesting questions.) But s
2.
How Offload Works: Inside the Rust CLI that sped up our tests by 6x(imbue.com)
12 points by thejash 5mo ago|0 comments
3.
by thejash 6mo ago
These are real problems, and I think you nailed it: the concentration of power is the core issue. I don't have a simple, perfect solution. We're just trying to make it possible for individuals and smaller companies to have access
4.
by thejash 6mo ago
(author here) I strongly agree that these systems start to break down once the code base gets larger (we've seen that with our own projects) But our reaction to it has been to say "ok, well the best practice in software engineerin
5.
by thejash 6mo ago
(author here) I agree that it's super important to understand what software actually does--that's part of the whole reason we made mngr in the first place! I believe we can use these types of tools to make software more understa
6.
A case study in testing with 100+ Claude agents in parallel(imbue.com)
68 points by thejash 6mo ago|54 comments
7.
Usefully run 100s of Claudes in parallel with mngr(imbue.com)
21 points by thejash 6mo ago|0 comments
8.
Software as Bonsai(substack.com)
2 points by thejash 8mo ago|0 comments
9.
AI should be Free Software(substack.com)
3 points by thejash 8mo ago|0 comments
10.
Why agents matter more than other AI(substack.com)
1 points by thejash 9mo ago|1 comments
11.
What's needed for AI agents to work(substack.com)
2 points by thejash 9mo ago|1 comments
12.
Decentralization or Dystopia: Pick 1(substack.com)
1 points by thejash 10mo ago|0 comments
13.
by thejash 1y ago
Since Sculptor allows you to use custom docker containers (devcontainers), you can check out the other projects in there. Then your primary project (from Sculptor's perspective) is simply whatever contains the devcontainer / doc
14.
by thejash 1y ago
It actually originally worked that way, and you can still mostly kinda use it that way (except that, because of CSRF protection, it's obnoxious -- run the program once to figure out the command line that the backend python process is s
15.
by thejash 1y ago
We have some docs here: https://github.com/imbue-ai/sculptor If you have any specific questions that aren't covered there, please let us know in Discord!
16.
by thejash 1y ago
Yup -- electron app, typescript / react on the front end, python on the backend
17.
by thejash 1y ago
I haven't tried it, but it might work if you set the env var yourself (I think you can create a `.env` file in `~/.sculptor/.env` and it will be injected into the environment for the agent) I'd give it a good 20% chanc
18.
by thejash 1y ago
Definitely! I'm very excited to get in support both for other coding agents, and for as many language models (and providers) as we can. Eventually what we want is for the whole thing to be open -- Sculptor, the coding agent, the under
19.
by thejash 1y ago
Nope, not based on vibekit, but it looks like a cool project! Our approach is a bit more custom and deeply integrated with the coding agents (ex: we understand when the turn has finished and can snapshot the docker container, allowing rollb
20.
by thejash 1y ago
Replying to each piece: > What if my application is not dockerized? Then claude runs in a container created from our default image, and any code it executes will run in that container as well. > Can Claude Code execute tests by itself
21.
by thejash 1y ago
Thanks! Please feel free to join discord if you run into any bugs or have any issues at all, we're happy to help: https://discord.gg/sBAVvHPUTE Suggestions welcome too!
22.
by thejash 1y ago
Since our launch 2 years ago, we've focused more on the "agents that code" part of our vision (so that everyone can make software) rather than the "training models from scratch" part (because there were so many good
23.
by thejash 1y ago
Our current plan is to make the source code available and make it free for personal use, but we're not quite ready to open-source it. Someday we'll probably have paid plans and business / enterprise licenses available as well
24.
Show HN: Sculptor – A UI for Claude Code(imbue.com)
176 points by thejash 1y ago|85 comments
25.
Sculptor: Catch and fix issues as you code(imbue.com)
61 points by thejash 1y ago|3 comments
26.
Carbs, a hyperparameter optimizer that scales small experiments to LLMs(imbue.com)
4 points by thejash 2y ago|1 comments
27.
by thejash 2y ago
We successfully scaled from a 7B run to a 70B run on the first try, with minimal training instability and no loss spikes. We also predicted performance of the 70B model based on experiment results from much smaller models. We accomplished t
28.
by thejash 2y ago
When training our 70B model, we sought to accurately evaluate models for natural language understanding and reasoning abilities. Surprisingly, we found that both open and closed models achieve nearly 100% accuracy when evaluated only on una
29.
Open-source, sanitized evaluation datasets for models that reason and code(imbue.com)
3 points by thejash 2y ago|1 comments
30.
Infrastructure setup and open-source scripts to train 70B model from bare metal(imbue.com)
325 points by thejash 2y ago|41 comments

More ›