babelfish

5 ms·

61.
by babelfish 4mo ago
When rights are equal, force wins. This is true for either the worker or the employer. Hence why employers frequently employed private firms to commit said violence on unions.
62.
by babelfish 4mo ago
What does this have to do with the steam deck?
63.
by babelfish 5mo ago
https://www.nytimes.com/interactive/2024/upshot/buy-rent-cal...
64.
by babelfish 5mo ago
Looks like the BMW i3 met a Magic Mouse
65.
by babelfish 5mo ago
So just use it for music. Who cares if the app bundle is bloated? If that's really your main criteria, just use the web player
66.
by babelfish 5mo ago
https://archive.is/ezljJ
67.
A Geometric Calculator Inside a Neural Network(goodfire.ai)
11 points by babelfish 5mo ago|0 comments
68.
by babelfish 5mo ago
they hire leading researchers, and leading researchers won't work for you unless they're able to publish
69.
by babelfish 5mo ago
I love both of these platforms. The Wikipedia "map" tab on the iOS app is also a great source of neat local oddities.
70.
The guide to RL environments: building and scaling them in the LLM era(huggingface.co)
3 points by babelfish 5mo ago|0 comments
71.
OpenAI Sells Statsig to Amplitude(amplitude.com)
5 points by babelfish 5mo ago|1 comments
72.
by babelfish 5mo ago
Wow, worse than Three Mile Island
73.
by babelfish 5mo ago
Source?
74.
by babelfish 6mo ago
I use Conductor which lets me flip trivially between OpenAI/Anthropic models
75.
by babelfish 6mo ago
It's data. Nobody is using Grok for SWE work, but they are using Cursor.
76.
by babelfish 6mo ago
Good on them to get $10B breakup terms, after the Twitter shitshow
77.
Former Pinterest team redesigns email with Extra – and it's good(techcrunch.com)
1 points by babelfish 6mo ago|0 comments
78.
by babelfish 6mo ago
Why would any media company care about what Objection says or agree to arbitration?
79.
by babelfish 6mo ago
No model card? No benchmarks? No usage examples? Nothing on the blog[0] since the acquisition? [0] https://x.ai/news
80.
GPT‑Rosalind for life sciences research(openai.com)
102 points by babelfish 6mo ago|30 comments
81.
by babelfish 6mo ago
Claude Code injects a 'warning: make sure this file isn't malware' message after every tool call by default. It seems like 4.7 is over-attending to this warning. @bcherny, filed a bug report feedback ID: 238e5f99-d6ee-45b5-98
82.
by babelfish 6mo ago
This is honestly pretty embarrassing for both parties. For OpenAI - it sounds like the CRO is trying to turn you into the Oracle or Salesforce of AI. For Anthropic - I hope your investors can see the actual revenue numbers.
83.
Sam Altman's home targeted in second attack(sfstandard.com)
156 points by babelfish 6mo ago|375 comments
84.
by babelfish 6mo ago
https://archive.is/eLssu
85.
by babelfish 6mo ago
Probably a better link: https://ai.meta.com/blog/introducing-muse-spark-msl/
86.
by babelfish 6mo ago
Totally. Best-in-class for SWE work (until Mythos gets released, if ever, but I suspect the rumored "Spud" will be out by then too)
87.
by babelfish 6mo ago
Combined results (Claude Mythos / Claude Opus 4.6 / GPT-5.4 / Gemini 3.1 Pro) SWE-bench Verified: 93.9% / 80.8% / — / 80.6% SWE-bench Pro: 77.8% / 53.4% / 57.7% / 54.
88.
Better MoE model inference with warp decode(cursor.com)
2 points by babelfish 6mo ago|0 comments
89.
by babelfish 6mo ago
Sure, for humans. Not sure they'll be the primary readers of code going forward
90.
by babelfish 6mo ago
Wow, maybe something is wrong with my setup. In Cursor 3, I am clicking "New Agent" at the top left. My root repository is correctly listed on top of the composer, and I clicked the icon to the right of it and selected 'Workt

More ›