transitivebs

8 ms·

61.
by transitivebs 3y ago
Consensus seems to be that it would be difficult for them to pull off, but this type of thing could be devastating for worldwide internet services in the future. The internet's decentralized nature helps a lot to mitigate this risk, bu
62.
by transitivebs 3y ago
what were the other DX flows you considered, and in your experimentation, what were the tradeoffs you found?
63.
Use Coffee to iterate on your frontend using AI(github.com)
15 points by transitivebs 3y ago|6 comments
64.
by transitivebs 3y ago
Coffee is a new OSS framework which lets you build and iterate on your frontend code – right from your own IDE. It's currently focused on React, and the team at Coframe behind the project has experimented with a few distinct DX flows f
65.
Cheat sheet on how to use LLMs effectively(twitter.com)
2 points by transitivebs 3y ago|0 comments
66.
Show HN: OpenOpenAI – OSS version of stateful Assistants API(github.com)
4 points by transitivebs 3y ago|0 comments
67.
by transitivebs 3y ago
Pretty sure OpenAI copyrighted the "GPT" suffix, so not sure what Kagi is naming an API this. I think OSS projects get a pass, but I'd imagine commercial competitors.
68.
by transitivebs 3y ago
https://archive.ph/Kn5xq
69.
by transitivebs 3y ago
Mangekyō
70.
Chidori – Declarative framework for AI agents (Rust, Python, and Node.js)(github.com)
153 points by transitivebs 3y ago|39 comments
71.
by transitivebs 3y ago
Friend of mine just open sourced his agent framework. It's written in Rust and supports some cool features like time-travel debugging, embedded code interpreter (Deno + Starlark), human-in-the-loop, and agent evals. Sharing here cause
72.
by transitivebs 3y ago
would be cool to see a loop of: js ⇒ python ⇒ js and then compare the output JS w/ the input JS. could get wilder too like: js ⇒ rust ⇒ typescript ⇒ java ⇒ js
73.
by transitivebs 3y ago
totally agreed that diff projects need to be careful w/ sharing proprietary code w/ third parties. openai's official stance is that it will never use API calls as training data, and that in my understanding it may retain API
74.
by transitivebs 3y ago
yep; it's generating unit tests and making sure that things are at least statically correct in terms of syntax. early days, but the ability to have loosely-defined acceptance criteria would go a long way here (like you would have for a
75.
by transitivebs 3y ago
agreed. I'm guessing the limiting factor here is training data w/ code from these languages that aren't as well represented in OSS
76.
by transitivebs 3y ago
this would be legendary. prolly going to fall on its face for something of this complexity, but "the next big thing always starts out looking like a toy"
77.
by transitivebs 3y ago
Great feedback. I'm not the author of this project, but in my understanding, it's the same as if you were to write the code yourself. The project doesn't publish anything and it works entirely locally aside from LLM calls (wh
78.
GPT-Migrate converts repos from one lang/framework to another(github.com)
213 points by transitivebs 3y ago|170 comments
79.
by transitivebs 3y ago
This looks absolutely amazing. The team building this is A+++ as well
80.
by transitivebs 3y ago
we're discussing this in https://www.chatgpthackers.dev discord
81.
by transitivebs 3y ago
I was also given a takedown notice by OpenAI for the ChatGPT twitter bot github repo: https://github.com/transitive-bullshit/chatgpt-twitter-bot This was ~2 months ago, and I'm fortunate enough to have a direct co
82.
by transitivebs 3y ago
This is a really great breakdown. Here's my super condensed advice on how to use LLMs effectively: https://twitter.com/transitive_bs/status/1643017583917174784
83.
by transitivebs 3y ago
Soooo many of our devs are going to take advantage of this: https://www.chatgpthackers.dev Looking forward to it!
84.
by transitivebs 3y ago
This is a dupe with a more sensational title of https://arxiv.org/abs/2304.03442
85.
by transitivebs 4y ago
I got access due to my open source contributions in this space. All you can do is apply and make your case as best you can.
86.
by transitivebs 4y ago
Hey Lee, Travis here. One piece of meta feedback for Vercel reading through all this is that whenever you have multiple ways of accomplishing something (like setting headers in vercel.json vs in the handlers themselves), it's bound to
87.
Show HN: Lex Fridman ChatGPT Plugin(twitter.com)
3 points by transitivebs 4y ago|2 comments
88.
A practical guide on how to use LLMs effectively(twitter.com)
1 points by transitivebs 4y ago|0 comments
89.
by transitivebs 4y ago
The example they gave of another company getting IDE Approval (Synchron) makes it seem possible, but I don't know how much of this is just general governmental inefficiencies and some level of systemic bias against Elon (stemming from
90.
by transitivebs 4y ago
Great analogy & breakdown. My go-to explanation is to think of ChatGPT like a really intelligent friend who's always available to help you out – but they're also super autistic, and you need to learn the best way to interact w

More ›