Content
@
0 reply
0 recast
0 reaction
Colin
@colin
I purchased a Claude team subscription for all /paragraph employees. I've been blown away by how good it is at coding and design. I can add essentially the entire Paragraph codebase to the context and ask targeted, specific questions about how to design or improve certain components.
11 replies
3 recasts
55 reactions
rish
@rish
We did the same for chatgpt earlier this year after the 4o release but wondering whether we should use Claude instead Honestly, seems like iOS / Android competition where they’ll keep flip flopping b/w which one’s the best at any given time
5 replies
0 recast
22 reactions
Will Papper
@will
The larger token context window makes a massive difference. I have a small script that concatenates all text in codebases or docs for uploading to Claude: https://gist.github.com/WillPapper/969751a6945c65e86e1069f59f81b3cd Would be impossible in GPT-4 due to the much smaller token context window
1 reply
0 recast
0 reaction
moreReese
@morereese
Do you use Raycast? The team subscription comes with tons of models for Raycast AI and also includes features like private extensions, shared quicklinks & commands, brand assets, etc I work mostly as a solopreneur and Raycast has been a 10x productivity tool for me. Would 100% get the teams sub if I ever built out and/or joined a small team https://www.raycast.com/teams
1 reply
0 recast
0 reaction
Colin
@colin
I'd recommend giving Claude a try. In my experience it's much better at certain coding tasks
1 reply
0 recast
1 reaction
🤖
@unbias
use cursor and add files to narrow context. otherwise i've use a custom context builder for large models that formats and/or describes files and their relationship. it works well as an ai first dev
0 reply
0 recast
0 reaction
Montez
@montez
Perplexity and Cursor setup seems to be a good hedge as the foundational models battle it out. Can’t wait until we get a standardized LLM profile to carry from app to app to reduce spending for the same models in different interfaces.
0 reply
0 recast
0 reaction
